New directions spur in-house research, industry outreach.
The war on terrorism has added a new sense of urgency to the Central Intelligence Agency’s science and technology development. The agency is accelerating its work in a number of key areas both to serve ongoing operations against al Qaida and to ensure long-term vigilance against asymmetric adversaries who are constantly changing their ways of operating.
With the onset of the war on September 11, technical programs were accelerated and new methodologies were introduced to support operation Enduring Freedom. In addition, companies developing new technologies for the Central Intelligence Agency (CIA) rushed prototype versions of their systems into action.
As the conflict unfolds, private industry’s role in intelligence technology development will grow, but in-house research and development will continue to serve a vital function. The commercial sector increasingly will serve as the font of innovation for secure communications systems, digital signal processing and information management software. In-house intelligence research will focus on developing vital application-specific breakthrough technologies such as new power sources, low-power electronics and new spectrum sensing. The underlying technical philosophy will be to view collection as a multifaceted task as well as a target-specific activity.
Dr. Donald M. Kerr, deputy director for science and technology at the CIA, states that his directorate’s fundamental element remains ensuring global access for the intelligence community’s collection needs. The key lies in selecting the approach that provides the highest probability of success, and this may involve employing different media in concert.
Kerr offers that one point has become even clearer since September 11: The kind of information the CIA needs to collect, process and disseminate is very different with the kinds of targets presented by al Qaida and other terrorist organizations. “To some degree, the premium shifts from a reconnaissance model—which has existed in the community for a long time—to a surveillance model, which has been more traditional in areas such as law enforcement,” he says. “In some sense, we are involved in a great manhunt.
“For the fleeting targets—whether they are physical or in the electronic domain—to assure yourself of the opportunity to capture those bits of information, surveillance seems to be the watchword.”
Kerr emphasizes that the directorate’s greatest priority in the midst of the war on terrorism is to continue supporting activities in the field. This support ranges from collection capability to providing personnel with the means to protect themselves. This activity is ongoing worldwide, not just in Afghanistan.
The three greatest challenges currently facing the CIA’s Directorate of Science and Technology existed to a large extent before September 11, although the attacks on that day may have emphasized them, Kerr notes. The first challenge involves the volume of intelligence data and the investments that must be made by the community to process it into a useful form of information. The increasing amounts of data available from more sources, coupled with improvements in collection capabilities, add up to a massive amount of information that must be processed into a useful product for the customer.
The tools used for this often are characterized as knowledge discovery or knowledge management, Kerr allows, and the need to process huge amounts of data in the post-September-11 era have brought this challenge into focus. Operation Enduring Freedom and its related operations have increased the urgency for solutions.
The second challenge, which is related to this problem, deals with the complexity of target sets. Kerr relates that intelligence now must go beyond simply determining what activities may be underway in a building, for example. Instead, the targets are becoming networks located in multiple geographic locations with a variety of transactional and communications connectivity—the very definition of a global terrorist organization.
“The intelligence community must focus on the mission of understanding that complex target rather than on the means of doing it,” Kerr emphasizes. The community must take advantage of all the different capabilities that it can marshal. “To steal a phrase from the [U.S.] Defense Department, we really must operate more and more as a ‘system of systems’ than has been the traditional approach,” he declares.
The issue of target set complexity can be addressed by using current collection techniques more effectively, Kerr notes. However, the directorate also is working to discover if other signatures, signals and areas of the spectrum can be exploited for vital information. This is where systems integration, especially the skills developed in industry over the past 20 years, can play a significant role.
The third challenge of the post-September-11 era is the ability to deal with discontinuities. These range from the collapse of the Soviet Union and end of the Warsaw Pact threat to the way enemies wage war.
One ongoing discontinuity is that organizations or individuals are changing their methods of operation to deter detection by intelligence assets. This problem is increasing as adversaries shift their communications from radio frequency transmissions to other means such as optical fiber.
Keeping up with discontinuities will require keeping abreast of technology advances taking place in industry and academia, Kerr allows. As new means of communication and modulation schemes are introduced, the intelligence community must be prepared to respond with its own technologies to enable continued surveillance.
With the intelligence community remaining a relatively small market in economic terms, it must take advantage of advances developed by industry, academia and other government organizations. The community always will be interested in outside developments in digital signal processing, Kerr states. The communications and information systems world has been transformed from analog to digital, and the intelligence community must be able both to use these digital data streams for its own purposes and to interact with them for information collection and processing.
Another area in which the community will rely heavily on private industry is secure communications. Kerr explains that the directorate will seek commercial developments that, with a modest amount of change, can be adapted for intelligence applications. This avoids the need for sizable investments in new technologies, he notes.
Software-defined radio is one technology that Kerr describes as “the most exciting in the longer term.” Special performance or unique encryption needs could be met through software modification instead of by adding modules or replacing hardware.
The directorate constantly keeps tabs on new areas of science or engineering for potential intelligence applications. Biotechnology, for example, is expanding into a wide range of commercial applications. The intelligence community is considering how biotechnology can play a role in its operations. It also is examining how the evolving technology can be turned into a threat against U.S. interests. A biotechnology center within the directorate keeps an eye on developments in industry and academia.
A related area of interest is biometrics. While public perception tends to concentrate on fingerprint identification and facial recognition, Kerr notes that this technology actually encompasses virtually any means of distinguishing one person from another.
Kerr states that the directorate will continue to focus its research and development on intelligence-unique technologies. These may range from national collection systems to relatively mundane items such as power sources, which are important in particular because they often are the key enablers for certain types of sensors, Kerr relates.
Low-power electronics may hold the key to future technology-based intelligence supremacy. Low-power units that also are small offer much greater flexibility in application, can be concealed more easily and consume less power. These types of circuitry also can enable smarter front-end sensors, which in turn can reduce bandwidth requirements.
A common misconception is that a sensor is designed to detect a final product—a chemical weapon, a biological agent or some other finished item. Kerr explains that the intelligence community needs sensors for every available signature ranging from a decision made by a national leader to the execution of that decision, possibly ending with production and delivery of a weapon.
The directorate is looking for systems that provide “the subtle indications” that relate to certain types of activities otherwise passing unnoticed. These activities might not take place in easily identifiable buildings or with readily noticed equipment, for example. Neither would communications or motorized traffic betray the activities.
Last month, the directorate sponsored an industry day at CIA headquarters. This differed from other similar events in that half of it took place at the unclassified level. Kerr explains that the goal was to attract firms not part of the traditional government contracting arena, especially small companies that might not realize how applicable their goods or services could be to the intelligence community. In addition to describing its needs at the unclassified level, the agency also provided information on how companies without an intelligence contracting history can do business with the community.
Three years ago, the agency created a venture capital organization to promote technology development in small companies. That technology incubation organization, In-Q-Tel (SIGNAL, April 2001, page 64), has helped insert important new technologies into intelligence operations.
“Post-9/11, In-Q-Tel has proved to be a considerable asset to both the CIA and the intelligence community at large,” Kerr declares. “Immediately [after the attacks], several companies within its portfolio made early versions of their products available for use both here at the CIA and at other agencies.” He describes these rapid product introductions as “mature beta tests” that, while not yet ready for the marketplace, still had enough functionality and utility for immediate use. “It was an incredible front door for us post-9/11,” he says. “A large number of companies, academic institutions and individuals volunteered their services or brought their ideas to us, and In-Q-Tel turned out to be a great portal.”
Kerr adds that In-Q-Tel also is serving as a go-between for introducing other technologies outside of its expertise to the appropriate elements of the intelligence community. The agency company often learns of budding technologies and business plans that are not within its purview but can be useful to other sectors of the community.
Kerr expects that In-Q-Tel’s focus is not likely to change in the near term. Its target area largely will remain the information technology business sector.
The community’s increased emphasis on human intelligence (HUMINT) is not coming at the expense of technical intelligence, Kerr maintains. “We are not a research and development organization. We are an operational organization, and we work hand in hand with our colleagues in the Directorate of Operations. It is a partnership because new technical capabilities enable case officers to be more effective and productive and generate a better HUMINT product.” He adds that one of the main missions in the science and technology directorate is to provide technical support to clandestine operations.
The directorate is working on pulling together multidisciplinary teams comprising several different –INTs to focus on specific problems. Kerr relates that this takes into account how many intelligence requirements cross traditional –INT definitions. The traditional requirements process, which typically is employed for acquiring expensive items, tends to hide the basic question that spurred the original acquisition request. “What we have been doing is using our all-source analysts within the community as the surrogate for our customer—the president and the national leadership—to be sure that some of the things that we are thinking of doing, in terms of adding to our technical tools for collection and analysis, still relate to their problem,” Kerr states. “The question is not how many bits per day were collected; the issue is whether the customer’s or the analyst’s question was answered properly.”
Kerr does not envision any significant reorganization of the science and technology directorate. “It has had several [reorganizations] in recent years, and I tend to think in terms of organizing to do the work. I don’t think that moving the boxes is the way to get people focused on our most pressing challenges.”
Funding is not a major problem, as Kerr states that Congress and the Bush administration “have treated us well” regarding financial resources. Recruiting and keeping talented people remain a challenge, even in light of the surge of applicants after September 11. The number of people seeking employment at the CIA increased tenfold in the wake of the attacks, but these people must be trained, mentored, integrated into the work force during wartime and retained over time. Kerr describes this as one of the greatest challenges facing the agency.