DID RUSSIAN STEAL STEALTH SECRET?

November 3, 2000

FEATURES & COMMENTARY

Seattle, WASH. — The Seattle Post-Intelligencer reports that a Russian mathematician who was given access to an American supercomputer loaded with stealth warplane design software is under investigation for espionage.

U.S. agents suspect that Aleksey Yeremin, who logged on to the supercomputer from Moscow, took advantage of Lockheed Martin and military security lapses to steal stealth technology secrets.

The 31/2-year investigation stretches from the heart of the old Soviet empire to Lockheed’s secretive Skunk Works plant in Southern California to a quiet suburb north of Seattle.

Yeremin, vice president of a software company based in Bothell, Wash. that did work for Lockheed, was e-mailed part of Lockheed’s modeling program for designing stealth planes. And, sources say, it would have been easy for him to steal the rest of it.

The potential loss is staggering: America’s global monopoly on radar invisibility.

In the spring of 1997, the FBI and the Air Force Office of Special Investigations, or OSI, launched the investigation. A short time later, Lockheed pulled the plug on its project with Yeremin. Federal criminal-justice sources say Yeremin, 46, has connections to the Russian military and now-defunct KGB.

It is unknown whether Yeremin got his hands on any classified information. But sources say that some of the unclassified information he did obtain should have been classified top-secret.

A retired Air Force three-star general said he is concerned that the apparent leak could help the Russians build their own stealth warplanes. “I don’t think what is at risk here is making these [U.S.] aircraft any more visible to radar,” said George Muellner, who played a key role in developing the nation’s stealth fleet. “But what is at risk is accelerating a country’s ability to develop and build these sorts of aircraft – to produce something that is a threat downstream,” he said. “If they started building and selling these things to the Iraqis, that would be a concern.”

The joint FBI-Air Force probe has so far yielded no arrests, and no one has been publicly charged with a crime.

Federal prosecutors, the FBI and the Air Force OSI refused to comment on the case. “What I can tell you is it is an ongoing investigation,” said Maj. Mike Richmond, an OSI spokesman at Andrews Air Force Base in Maryland.

Working out of his Bothell home, high-tech entrepreneur Russ Sarbora plucked Yeremin from a Russian military aircraft-design center a decade ago and collaborated with him to start a company called Elegant Mathematics.

Today, Sarbora is shocked that his business partner is under investigation in an espionage case. “I believe he is not an agent for any foreign power,” Sarbora said. “It would surprise me enormously if he turned out to be.”

Yeremin “readily acknowledged working for the Soviet military,” according to Sarbora, but “prior to 1990, if you were a scientist of any stature, you worked for the Soviet military structure. You didn’t get to work, or support in school without working for the military.”

Yeremin and his associates in Russia were working on computers loaded with software known as MM3D, or Method of Moments in Three Dimensions. It would have been easy for people with their expertise to steal the portions of MM3D they hadn’t already been given, the former Lockheed computer expert said.

And it wasn’t just MM3D that was at risk of being stolen. Also loaded on the expert’s home computer were test fixtures – secret computer representations of stealth aircraft structures. The expert said he had been assured by Lockheed engineers that the test fixtures were not classified. But after the investigation began, federal agents claimed some were top-secret. One of the computers Yeremin was given access to also contained data on the performance characteristics of radar-absorbing materials that coat stealth planes, the expert said. Those data, however, did not specifically identify the name or chemical composition of the materials, he said.

Gen. Muellner, now vice president-general manager of The Boeing Co.’s Phantom Works research and development division, cautions that there may be no way of telling if the apparent leak is devastating or merely distressing.

Like Lee and former CIA Director John Deutch, who is under investigation for having classified information on his home computer, the Lockheed computer expert was working on an unsecure computer. His home computer was loaded with MM3D. The program, he said, contains a feature called an “optimizer” by which designers can simulate changes in aircraft configuration or materials and quickly see how that affects the plane’s radar invisibility.

Unlike in the Deutch case, the computer expert said he had written permission from Lockheed’s security and legal departments to work on the program at home. Only after the investigation began was he told that the optimizer should have been classified top-secret, he said.

The computer expert said the reason he was allowed to work at home is that at night, he could access government and private supercomputers for free, saving Lockheed the expense of maintaining an in-house supercomputer on which he could work. “I really campaigned hard to get an in-house supercomputer so I wouldn’t have to go out on the Net,” he said.

He said he repeatedly sought guidance from Lockheed security officials over how far he could go with Yeremin but was either ignored or rebuffed. “I had continually apprised security of everything,” he said. “Somebody out there knew what Alex’s background was. But somehow the communication to the (security) guy that was supposed to be covering my back didn’t happen.”

A former Lockheed security official agreed. “It was bungled by Lockheed and the Air Force. From what I saw, everybody kind of snoozed through, kind of kissed it off. It was just ‘keep an eye on it and give a report once in a while,'” said the former official, who spoke on condition of anonymity.

The computer expert said he left Lockheed after the company pulled his access to work on the stealth program. He remained a Lockheed consultant for a time, then left for another job. He maintains that any information he gave Yeremin “pales in comparison” to open articles written by scientists and engineers, and technical information available over the Internet. “The bigger issue is that these guys (Russian mathematicians) had access to a lot of computers and computer networks,” the former employee said. “Their software was being run on many American networks and supercomputers.”

The 48-year-old Sarbora is a longtime software programmer and former computer industry executive. He moved to Seattle in 1988 to become vice president of quality assurance and technical support at Microrim Inc., a database software company.

During the Goodwill Games of 1990, Sarbora volunteered to work on a project showcasing the best of Soviet technology and met several Russian scientists. The following year, Sarbora went to Russia on a business trip, looking to bring back software he could sell to U.S. companies. He said he met Yeremin at the Central Aerohydrodynamics Institute, about 30 miles southeast of Moscow, where Russian warplanes are designed.

It was shortly after the fall of the Soviet Union. Work for scientists and mathematicians in Russia was rapidly disappearing. The U.S. government feared they would peddle their expertise in “rogue” states like Libya or Iraq.

When Sarbora learned that Yeremin had created an algorithm, or calculation method, that could achieve on relatively primitive computers what U.S. scientists did on supercomputers, he immediately grasped the scientific and commercial potential.

Yeremin joined forces with Sarbora to launch Elegant Mathematics. They incorporated in Washington state with Sarbora as president and Yeremin as vice president. The main offices, however, were thousands of miles away in Moscow, at the Russian Academy of Sciences.

At its peak, the company employed about 20 mathematicians, physicists and software experts from the Russian academy, Steklov Mathematical Institute and Moscow State University.

“We insisted on maintaining the team in Russia,” Sarbora said. “The idea is that after the transition from communism to capitalism, there would be a few teams that could maintain mathematics in Russia.”

“We developed technology that would reduce computational costs of solving problems in stealth technology by reducing the number of calculations by an order of 10,” Sarbora said.

Sarbora said his company scored its first major contract with Cray, a supercomputer maker, and continued working on its software-development project at the Pittsburgh Supercomputing Center and at IBM.

In 1994, Elegant Mathematics was hired by Lockheed to improve the efficiency and speed of computer simulations related to the interaction between the aircraft and radar. That effectively meant infusing MM3D with Yeremin’s “solver” program.

The computer expert’s job at Lockheed involved using computers to simulate how a radar wave reflects off the surface of a plane. The whole idea of stealth technology is to prevent that reflection so the wave echo doesn’t return to the radar antenna and get read on a radar screen.

Over the next three years, the computer expert said, he met with Yeremin at Skunk Works at least 15 times. The sprawling facility in the Mojave Desert is surrounded by chain-link fences topped with razor wire. Armed guards are stationed at the gates. Access to secure buildings requires personal security codes and badges that open electronic locks.

The former Lockheed security official said the company lusted after Elegant Mathematics’ promised cost-cutting technology. “That was where the greed came in. Yeremin was offering this tantalizing carrot in front of everybody.”

“He’s an interesting character,” the computer expert said of Yeremin. “I always liked him a lot because of his zeal for the task at hand. But he leaves a trail of people very pissed off because he’s so arrogant. He’s brilliant, but not as brilliant as he thinks he is.”

The expert, however, said he believed in Yeremin’s algorithm. “I have seen it solve big problems,” he said. “Seeing is believing. I don’t think they were peddling snake oil.”

Others disagree. Even before the security concerns were raised, Yeremin’s breakthrough was being officially snubbed. Elegant Mathematics’ grant application to continue research into its algorithm failed to pass review at the Defense Advanced Research Projects Agency, a Defense Department agency that sponsors exotic research.

“The methods he (Yeremin) has developed are not considered competitive in this country. He got attention for a while, until his balloon was punctured,” said University of Illinois professor Eric Michielssen, an expert in the use of computer modeling of radar problems.

The computer expert said he sent Yeremin sections of source code, the underlying components of the MM3D program, by e-mail in early 1997. He said the source code was not classified, and the portions he sent were not related to the physics of stealth, but to technical requirements of inputting and retrieving data that Yeremin needed for his work.

The expert was testing Yeremin’s software on supercomputers at the NASA Ames Research Center in California, the Oak Ridge National Laboratory in Tennessee and at IBM in New York. The entire MM3D program was loaded on the NASA Ames and IBM computers, as well as on his home computer, he said.

Federal agents told the expert there were worries over whether Yeremin’s software might have been implanted with the hacking capability of e-mailing information from those computers back to Russia. And Yeremin had direct, online access to the IBM supercomputer, the expert said.

The expert said the agents “were never 100 percent sure that I had not sent the whole (MM3D) source code to Yeremin. I could have sent the whole farm to him. Which, of course, I did not.”

There was also the concern that Yeremin or one of his Russian colleagues could have hacked into the computer expert’s home computer. The expert said he followed the rules by reporting his e-mail exchanges with Yeremin to Lockheed security. Security people later turned over to Air Force OSI agents information on the e-mails, triggering the investigation.

What the Air Force calls a “Red Team” – a group of technology, military and security experts – was mobilized to assess the potential damage to national security. One inside source said the team called it “catastrophic.” Other investigators are not convinced that Yeremin was involved in espionage, federal sources said.

Like Muellner, the retired general, the former Lockheed computer expert worries that the Russians might have stolen technology that gives them a big boost in the design and construction of stealth aircraft.

“We don’t know if he got the source codes,” the computer expert said. “That was the supposition due to the fact that we were working on the same computer. A routine hacker with the kind of access Yeremin had could have gotten the codes.”

When federal agents came calling on the computer expert on June 23, 1997, he was aghast. They entered his home in California’s Tehachapi Mountains and demanded that he turn over his computer. He said he cooperated fully. They told him he may have seriously damaged national security. “I was traumatized beyond belief,” he said. That was more than three years ago.

The agents still have his hard drive. He hasn’t been arrested since then, he asserts, because he didn’t deliver classified information to Yeremin, had permission to work on his home computer and had no intent to damage national security.

The FBI paid a similar call on Sarbora in 1997. Sarbora says it marked the beginning of the end for Elegant Mathematics.

Lockheed soon cut its ties to the company. The specter of an espionage investigation scared off almost everyone else.

The FBI, Sarbora said, interviewed “our customers and prospective customers; anyone we had a relationship with. That had a very chilling effect on business. It pretty much put a box around it and shut it down.”

In 1997 and 1998, Sarbora was questioned several times by the FBI. So was Yeremin, who has not been seen in the United States since 1999. While Yeremin is still considered a “person of interest” in the spy case, Sarbora said the agents ultimately told him he was no longer under suspicion.

By then, it was too late. The dreams of Yeremin and Sarbora had been dashed. So were the hopes of the crack team of Russian scientists. They “are all scratching for jobs,” Sarbora said. “Elegant was on life-support – comatose – and remains so.”

============================================================

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire