Privacy Oriented

After a little hiatus, I’ve returned to this blog. Some real content is soon to come.

Terror Fight Blurs Line Over Domain; Tracking Email

WASHINGTON, D.C. — Five years ago, Congress killed an experimental Pentagon antiterrorism program meant to vacuum up electronic data about people in the U.S. to search for suspicious patterns. Opponents called it too broad an intrusion on Americans’ privacy, even after the Sept. 11 terrorist attacks.

But the data-sifting effort didn’t disappear. The National Security Agency, once confined to foreign surveillance, has been building essentially the same system.

The central role the NSA has come to occupy in domestic intelligence gathering has never been publicly disclosed. But an inquiry reveals that its efforts have evolved to reach more broadly into data about people’s communications, travel and finances in the U.S. than the domestic surveillance programs brought to light since the 2001 terrorist attacks.

Congress now is hotly debating domestic spying powers under the main law governing U.S. surveillance aimed at foreign threats. An expansion of those powers expired last month and awaits renewal, which could be voted on in the House of Representatives this week. The biggest point of contention over the law, the Foreign Intelligence Surveillance Act, is whether telecommunications and other companies should be made immune from liability for assisting government surveillance.

Largely missing from the public discussion is the role of the highly secretive NSA in analyzing that data, collected through little-known arrangements that can blur the lines between domestic and foreign intelligence gathering. Supporters say the NSA is serving as a key bulwark against foreign terrorists and that it would be reckless to constrain the agency’s mission. The NSA says it is scrupulously following all applicable laws and that it keeps Congress fully informed of its activities.

According to current and former intelligence officials, the spy agency now monitors huge volumes of records of domestic emails and Internet searches as well as bank transfers, credit-card transactions, travel and telephone records. The NSA receives this so-called “transactional” data from other agencies or private companies, and its sophisticated software programs analyze the various transactions for suspicious patterns. Then they spit out leads to be explored by counterterrorism programs across the U.S. government, such as the NSA’s own Terrorist Surveillance Program, formed to intercept phone calls and emails between the U.S. and overseas without a judge’s approval when a link to al Qaeda is suspected.

The NSA’s enterprise involves a cluster of powerful intelligence-gathering programs, all of which sparked civil-liberties complaints when they came to light. They include a Federal Bureau of Investigation program to track telecommunications data once known as Carnivore, now called the Digital Collection System, and a U.S. arrangement with the world’s main international banking clearinghouse to track money movements.

The effort also ties into data from an ad-hoc collection of so-called “black programs” whose existence is undisclosed, the current and former officials say. Many of the programs in various agencies began years before the 9/11 attacks but have since been given greater reach. Among them, current and former intelligence officials say, is a longstanding Treasury Department program to collect individual financial data including wire transfers and credit-card transactions.

It isn’t clear how many of the different kinds of data are combined and analyzed together in one database by the NSA. An intelligence official said the agency’s work links to about a dozen antiterror programs in all.

A number of NSA employees have expressed concerns that the agency may be overstepping its authority by veering into domestic surveillance. And the constitutional question of whether the government can examine such a large array of information without violating an individual’s reasonable expectation of privacy “has never really been resolved,” said Suzanne Spaulding, a national-security lawyer who has worked for both parties on Capitol Hill.

NSA officials say the agency’s own investigations remain focused only on foreign threats, but it’s increasingly difficult to distinguish between domestic and international communications in a digital era, so they need to sweep up more information.

The Fourth Amendment

In response to the Sept. 11 attacks, then NSA-chief Gen. Michael Hayden has said he used his authority to expand the NSA’s capabilities under a 1981 executive order governing the agency. Another presidential order issued shortly after the attacks, the text of which is classified, opened the door for the NSA to incorporate more domestic data in its searches, one senior intelligence official said.

The NSA “strictly follows laws and regulations designed to preserve every American’s privacy rights under the Fourth Amendment to the U.S. Constitution,” agency spokeswoman Judith Emmel said in a statement, referring to the protection against unreasonable searches and seizures. The Office of the Director of National Intelligence, which oversees the NSA in conjunction with the Pentagon, added in a statement that intelligence agencies operate “within an extensive legal and policy framework” and inform Congress of their activities “as required by the law.” It pointed out that the 9/11 Commission recommended in 2004 that intelligence agencies analyze “all relevant sources of information” and share their databases.

Two former officials familiar with the data-sifting efforts said they work by starting with some sort of lead, like a phone number or Internet address. In partnership with the FBI, the systems then can track all domestic and foreign transactions of people associated with that item — and then the people who associated with them, and so on, casting a gradually wider net. An intelligence official described more of a rapid-response effect: If a person suspected of terrorist connections is believed to be in a U.S. city — for instance, Detroit, a community with a high concentration of Muslim Americans — the government’s spy systems may be directed to collect and analyze all electronic communications into and out of the city.

The haul can include records of phone calls, email headers and destinations, data on financial transactions and records of Internet browsing. The system also would collect information about other people, including those in the U.S., who communicated with people in Detroit.

The information doesn’t generally include the contents of conversations or emails. But it can give such transactional information as a cellphone’s location, whom a person is calling, and what Web sites he or she is visiting. For an email, the data haul can include the identities of the sender and recipient and the subject line, but not the content of the message.

Intelligence agencies have used administrative subpoenas issued by the FBI — which don’t need a judge’s signature — to collect and analyze such data, current and former intelligence officials said. If that data provided “reasonable suspicion” that a person, whether foreign or from the U.S., was linked to al Qaeda, intelligence officers could eavesdrop under the NSA’s Terrorist Surveillance Program.

The White House wants to give companies that assist government surveillance immunity from lawsuits alleging an invasion of privacy, but Democrats in Congress have been blocking it. The Terrorist Surveillance Program has spurred 38 lawsuits against companies. Current and former intelligence officials say telecom companies’ concern comes chiefly because they are giving the government unlimited access to a copy of the flow of communications, through a network of switches at U.S. telecommunications hubs that duplicate all the data running through it. It isn’t clear whether the government or telecom companies control the switches, but companies process some of the data for the NSA, the current and former officials say.

On Friday, the House Energy and Commerce Committee released a letter warning colleagues to look more deeply into how telecommunications data are being accessed, citing an allegation by the head of a New York-based computer security firm that a wireless carrier that hired him was giving unfettered access to data to an entity called “Quantico Circuit.” Quantico is a Marine base that houses the FBI Academy; senior FBI official Anthony DiClemente said the bureau “does not have ‘unfettered access’ to any communication provider’s network.”

The political debate over the telecom information comes as intelligence agencies seek to change traditional definitions of how to balance privacy rights against investigative needs. Donald Kerr, the deputy director of national intelligence, told a conference of intelligence officials in October that the government needs new rules. Since many people routinely post details of their lives on social-networking sites such as MySpace, he said, their identity shouldn’t need the same protection as in the past. Instead, only their “essential privacy,” or “what they would wish to protect about their lives and affairs,” should be veiled, he said, without providing examples.

Social-Network Analysis

The NSA uses its own high-powered version of social-network analysis to search for possible new patterns and links to terrorism. The Pentagon’s experimental Total Information Awareness program, later renamed Terrorism Information Awareness, was an early research effort on the same concept, designed to bring together and analyze as much and as many varied kinds of data as possible. Congress eliminated funding for the program in 2003 before it began operating. But it permitted some of the research to continue and TIA technology to be used for foreign surveillance.

Some of it was shifted to the NSA — which also is funded by the Pentagon — and put in the so-called black budget, where it would receive less scrutiny and bolster other data-sifting efforts, current and former intelligence officials said. “When it got taken apart, it didn’t get thrown away,” says a former top government official familiar with the TIA program.

Two current officials also said the NSA’s current combination of programs now largely mirrors the former TIA project. But the NSA offers less privacy protection. TIA developers researched ways to limit the use of the system for broad searches of individuals’ data, such as requiring intelligence officers to get leads from other sources first. The NSA effort lacks those controls, as well as controls that it developed in the 1990s for an earlier data-sweeping attempt.

Sen. Ron Wyden, an Oregon Democrat and member of the Senate Intelligence Committee who led the charge to kill TIA, says “the administration is trying to bring as much of the philosophy of operation Total Information Awareness as it can into the programs they’re using today.” The issue has been overshadowed by the fight over telecoms’ immunity, he said. “There’s not been as much discussion in the Congress as there ought to be.”

Opportunity for Debate

But Sen. Kit Bond of Missouri, the ranking Republican on the committee, said by email his committee colleagues have had “ample opportunity for debate” behind closed doors and that each intelligence program has specific legal authorization and oversight. He cautioned against seeing a group of intelligence programs as “a mythical ‘big brother’ program,” adding, “that’s not what is happening today.”

The legality of data-sweeping relies largely on the government’s interpretation of a 1979 Supreme Court ruling allowing records of phone calls — but not actual conversations — to be collected without a judge issuing a warrant. Multiple laws require a court order for so-called “transactional’” records of electronic communications, but the 2001 Patriot Act lowered the standard for such an order in some cases, and in others made records accessible using FBI administrative subpoenas called “national security letters.” (Read the ruling.)

A debate is brewing among legal and technology scholars over whether there should be privacy protections when a wide variety of transactional data are brought together to paint what is essentially a profile of an individual’s behavior. “You know everything I’m doing, you know what happened, and you haven’t listened to any of the contents” of the communications, said Susan Landau, co-author of a book on electronic privacy and a senior engineer at Sun Microsystems Laboratories. “Transactional information is remarkably revelatory.”

Ms. Spaulding, the national-security lawyer, said it’s “extremely questionable” to assume Americans don’t have a reasonable expectation of privacy for data such as the subject-header of an email or a Web address from an Internet search, because those are more like the content of a communication than a phone number. “These are questions that require discussion and debate,” she said. “This is one of the problems with doing it all in secret.”

Gen. Hayden, the former NSA chief and now Central Intelligence Agency director, in January 2006 publicly defended the activities of the Terrorist Surveillance Program after it was disclosed by the New York Times. He said it was “not a driftnet over Lackawanna or Fremont or Dearborn, grabbing all communications and then sifting them out.” Rather, he said, it was carefully targeted at terrorists. However, some intelligence officials now say the broader NSA effort amounts to a driftnet. A portion of the activity, the NSA’s access to domestic phone records, was disclosed by a USA Today article in 2006.

The NSA, which President Truman created in 1952 through a classified presidential order to be America’s ears abroad, has for decades been the country’s largest and most secretive intelligence agency. The order confined NSA spying to “foreign governments,” and during the Cold War the NSA developed a reputation as the world’s premier code-breaking operation. But in the 1970s, the NSA and other intelligence agencies were found to be using their spy tools to monitor Americans for political purposes. That led to the original FISA legislation in 1978, which included an explicit ban on the NSA eavesdropping in the U.S. without a warrant.

Big advances in telecommunications and database technology led to unprecedented data-collection efforts in the 1990s. One was the FBI’s Carnivore program, which raised fears when it was in disclosed in 2000 that it might collect telecommunications information about law-abiding individuals. But the ground shifted after 9/11. Requests for analysis of any data that might hint at terrorist activity flooded from the White House and other agencies into NSA’s Fort Meade, Md., headquarters outside Washington, D.C., one former NSA official recalls. At the time, “We’re scrambling, trying to find any piece of data we can to find the answers,” the official said.

The 2002 congressional inquiry into the 9/11 attacks criticized the NSA for holding back information, which NSA officials said they were doing to protect the privacy of U.S. citizens. “NSA did not want to be perceived as targeting individuals in the United States” and considered such surveillance the FBI’s job, the inquiry concluded.

FBI-NSA Projects

The NSA quietly redefined its role. Joint FBI-NSA projects “expanded exponentially,” said Jack Cloonan, a longtime FBI veteran who investigated al Qaeda. He pointed to national-security letter requests: They rose from 8,500 in 2000 to 47,000 in 2005, according to a Justice Department inspector general’s report last year. It also said the letters permitted the potentially illegal collection of thousands of records of people in the U.S. from 2003-05. Last Wednesday, FBI Director Robert Mueller said the bureau had found additional instances in 2006.

It isn’t known how many Americans’ data have been swept into the NSA’s systems. The Treasury, for instance, built its database “to look at all the world’s financial transactions” and gave the NSA access to it about 15 years ago, said a former NSA official. The data include domestic and international money flows between bank accounts and credit-card information, according to current and former intelligence officials.

The NSA receives from Treasury weekly batches of this data and adds it to a database at its headquarters. Prior to 9/11, the database was used to pursue specific leads, but afterward, the effort was expanded to hunt for suspicious patterns.

Through the Treasury, the NSA also can access the database of the Society for Worldwide Interbank Financial Telecommunication, or Swift, the Belgium-based clearinghouse for records of international transactions between financial institutions, current and former officials said. The U.S. acknowledged in 2006 that the CIA and Treasury had access to Swift’s database, but said the NSA’s Terrorism Surveillance Program was separate and that the NSA provided only “technical assistance.” A Treasury spokesman said the agency had no comment.

Through the Department of Homeland Security, airline passenger data also are accessed and analyzed for suspicious patterns, such as five unrelated people who repeatedly fly together, current and former intelligence officials said. Homeland Security shares information with other agencies only “on a limited basis,” spokesman Russ Knocke said.

NSA gets access to the flow of data from telecommunications switches through the FBI, according to current and former officials. It also has a partnership with FBI’s Digital Collection system, providing access to Internet providers and other companies. The existence of a shadow hub to copy information about AT&T Corp. telecommunications in San Francisco is alleged in a lawsuit against AT&T filed by the civil-liberties group Electronic Frontier Foundation, based on documents provided by a former AT&T official. In that lawsuit, a former technology adviser to the Federal Communications Commission says in a sworn declaration that there could be 15 to 20 such operations around the country. Current and former intelligence officials confirmed a domestic network of hubs, but didn’t know the number. “As a matter of policy and law, we can not discuss matters that are classified,” said FBI spokesman John Miller.

The budget for the NSA’s data-sifting effort is classified, but one official estimated it surpasses $1 billion. The FBI is requesting to nearly double the budget for the Digital Collection System in 2009, compared with last year, requesting $42 million. “Not only do demands for information continue to increase, but also the requirement to facilitate information sharing does,” says a budget justification document, noting an “expansion of electronic surveillance activity in frequency, sophistication, and linguistic needs.”

New-ish privacy threats have emerged, however unlikely, and this includes a threat to your encrypted information. From Popular Mechanics’ April ‘08 Issue:

“Individuals, companies and federal agencies could all be at risk from foreign governments or criminal enterprises. A computer chip built with a subtle error might allow an identity-theft ring to hack past the encryption used to connect customers with their banks. Flash memory hidden inside a corporation’s networked printers could save an image file of every document it printed, then send out the information.”

The entire article is reprinted below, for archiving’s sake:

This past January, two brothers from Texas, Michael and Robert Edman, appeared in court to face federal charges of selling counterfeit computer equipment to, among others, the Air Force, Marine Corps, Federal Aviation Administration, Department of Energy, numerous universities and defense contractors such as Lockheed Martin. According to prosecutors, the pair, working largely out of Michael Edman’s house in the rural town of Richmond, bought cheap network cards from a supplier in China. They also purchased labels and boxes carrying the logo of Cisco Systems, the U.S.-based hardware giant. Until a source in China tipped off the FBI, no one could tell that the parts were Cisco knockoffs rather than the real thing.

An attorney for the Edmans says that they, too, were victims—duped by overseas suppliers. But one thing is clear: The case is about a lot more than trademark infringement. Security experts warn that as supply chains become more global and more opaque, no one can be sure what parts are going into the computers that run, well, everything—from air traffic control towers to banks to weapons systems. Secretary of Homeland Security Michael Chertoff raised the issue recently at a briefing attended by Popular Mechanics and others. “Increasingly when you buy computers they have components that originate … all around the world,” he said. “We need to look at … how we assure that people are not embedding in very small components … that can be triggered remotely.”

Software vulnerabilities and online scams receive plenty of public attention. Viruses, Trojan horses, spyware, phishing schemes that trick people into providing financial data—all have made headlines in recent years. The emerging hardware threat is different. Imagine buying a computer, printer, monitor, router or other device in which malevolent instructions, or at least security loopholes, are etched permanently into the silicon.

Individuals, companies and federal agencies could all be at risk from foreign governments or criminal enterprises. A computer chip built with a subtle error might allow an identity-theft ring to hack past the encryption used to connect customers with their banks. Flash memory hidden inside a corporation’s networked printers could save an image file of every document it printed, then send out the information. In a disturbing national-security scenario, overseas agents might be able to hard-wire instructions to bring down a Department of Defense system on a predetermined date or in response to an external trigger. In the time it took to bring the systems back online, a military assault could be underway.

Shadowy Threat

When a software problem is detected, thousands or millions of computers can be fixed within hours with a software patch. Discover a malevolent hardware component, however, and machines need to be fixed one by one by one. On a large network it could take months—if the problem were detected at all.

“There are a whole bunch of functions inside each chip that you have no direct access to,” says Stephen Kent, chief information security scientist for BBN Technologies and a member of the Intelligence Science Board, which advises U.S. intelligence agencies. “We passed the point a long time ago when you could combinatorially test all the possible inputs for a complex chip. If somebody hid a function that, given the right inputs, could cause the chip to do something surprising, it’s not clear how you could test for that.”

Such tampering wouldn’t have to occur in a factory where computer components were built. In fact, repair businesses and subcontractors may pose a greater danger. “A skilled and capable adversary could replace a chip on a circuit board with a very similar one,” says John Pironti, a security expert for information technology consulting firm Getronics. “But this chip would have malicious instructions added to the programming.” The strategy wouldn’t be practical for running a broad identity-theft operation, but it might allow spies to focus an attack on a valuable corporate or government target—gaining access to equipment, then doctoring it with hidden functions.

However, not all experts agree that the risk is severe. After all, there’s never been a report of a foreign country or criminal outfit using such technology to steal information or commit sabotage. (The United States did successfully conduct such a mission against the Soviet Union during the Cold War.)

“It’s certainly possible for the world’s major espionage services to secretly plant vulnerabilities in our microprocessors, but the threat is overblown,” says Bruce Schneier, chief technology officer of the data security company BT Counterpane. “Why would anyone go through the effort and take the risk, when there are thousands of vulnerabilities in our computers, networks and operating systems waiting to be discovered with only a few hours’ work?”

The National Security Agency and Defense Department aren’t convinced. There’s no way to know if they are reacting to an imminent danger or simply swinging at shadows, but security professionals are scrambling to guard their electronics supply chains.

Building Chips in China

In September 2007, Intel broke ground on “Fab 68,” a silicon-wafer fabricating plant in Dalian, China. The plant is Intel’s first chip manufacturing facility in China, but the company already operates facilities for testing, as well as research and development, all over the world, from India to Costa Rica to Russia. Rival AMD is planning to build a fab in India. Several other American chipmakers, including Applied Materials and National Semiconductor, have facilities in China. In all, less than 25 percent of the world’s chipmaking capacity is still located within the United States.

The companies that move offshore are trying to stay competitive in commercial markets. As a side effect of globalization, however, the Defense Department is finding itself with fewer domestic sources of the specialized chips—often outdated by Best Buy standards—that help run weapons platforms that range from advanced aircraft to missile guidance systems. These are the electronic components that might pose the most inviting target for a foreign power.

The NSA is trying to counter the threat with a program called Trusted Foundry Access that accredits companies that supply specialized electronics to government agencies. Ten companies have joined the program since 2004—the inaugural deal, with IBM, cost the government a reported $600 million. To participate, manufacturers need to take measures such as obtaining security clearances for staff members and quarantining computer design tools from the Internet. Further, “The facilities must be on-shore or in a closely allied country,” says a Defense Department official involved with the program.

One potential flaw in the program is that it covers “just a slice of the life cycle,” says Jim Gosler, a Sandia National Laboratories researcher who has spent time probing U.S. electronics systems to identify vulnerabilities. “You have to make sure the component stays trusted—they get out and about” once the equipment leaves the factory and goes into service.

More critically, even well-funded initiatives can’t permanently withstand the forces pushing microchip production offshore. Ultimately, trying too hard to isolate American chip-making might simply help foreign-owned chip manufacturers challenge U.S. dominance in the field. “It’s a pretty hairy situation to look out 10 or 15 years and have to ask, ‘Where are we going to get our technology?’” the Defense official says.

DARPA, the Defense Advanced Research Projects Agency, does have another plan. Through a new initiative called Trust in ICs (microchips are also called integrated circuits or ICs), the agency has contracted with Raytheon, MIT, Johns Hopkins University and others to find ways to protect chips from tampering and to detect vulnerabilities if they do occur.

Ultimately, though, chips may be too complex to secure completely. “Even if you found something, you could never be confident you found everything,” Gosler says. “That’s the awful nature of this business.”