Address By John C. Gannon
Chairman, National Intelligence Council
to the Washington College of Law
at American University, Washington, DC
October 6, 2000
Thank you. It is a particular pleasure to be here this morning to address
the Washington College of Law on the subject of the "Strategic
Use of Open-Source Information." The open-source world represents
a major challenge to the US Intelligence Community, which is, in addition
to being an espionage service, is one of the world’s biggest information-based
businesses. The open-source challenge is a longstanding high priority
for us, and our response to it is very much a dynamic work in progress.
Let me begin by saying that open-source information is not what it use to be. When I was working as a regional analyst over a decade ago, "open source" largely meant information from foreign newspapers and the electronic media, which was collected mostly by the Foreign Broadcast Information Service (FBIS). "Open source" was "frosting on the cake" of source material dominated by clandestine collection, SIGINT, IMINT, and HUMINT.
Today, open source has expanded well beyond "frosting" deep into the cake. It is indispensable to the production of authoritative analysis. It increasingly contains the best information to answer our most important questions. It is not just media reports, which are now just a small, decreasing piece of the open-source pie. It is a vast array of documents and reports, which are publicly retrievable but, nonetheless, often hard to retrieve from today’s high-volume, high-speed information flow. And it is vital unrecorded information in the heads of knowledgeable people we need to engage. Accessing open-source information, in short, is a multi-faceted challenge that can only be met with a multi-front response or strategy.
Let me begin by trying to encapsulate my remarks in five key points I would like to make this morning:
First, open-source information today is more important than ever in the post-Cold War world, in which intelligence targets are more diverse in complexity and more dispersed in geography. Closed societies in the Former Soviet Union and in Eastern Europe have opened up, and reliable information now proliferates. The revolution in information technology, at the same time, has vastly increased the volume and speed of the information flow across the globe and across our computer screens. Open-source information now dominates the universe of the intelligence analyst, and this is unlikely to change for the foreseeable future.
Second, technology is a major part of the answer but it is no substitute for the other essential part, people. To deal with the open-source challenge, the Intelligence Community must invest more in technology to give us the analytical tools we need to access and exploit the vast information available to us, and in our people on whose expertise we must rely more than ever to prioritize and interpret this information. My experience over the past decade has taught me that the greater the volume of information to assess, the stronger must be the expertise we bring to the effort to evaluate it.
Third, we also need to have close and enduring partnerships in the commercial world to benefit from the private sector’s continuing pursuit of new technology and from its best practices in dealing with the open-source challenge. I will tell you about two new organizations CIA has developed to build and sustain such outside partnerships, the Office of Advanced Analytic Tools (AAT) and In-Q-Tel. The IC, by itself, simply cannot stay ahead of the technological curve and it knows it.
Fourth, we must continue to change the attitude and behavior of our analysts toward the outside world. We must provide incentives for our analysts to get out from behind their desks to engage with substantive experts and other outside sources of useful—and increasingly critical—information that cannot be captured by clandestine collectors or traditional open-source collectors such as the Foreign Broadcast Information Service (FBIS). This is an imperative, not an option for us.
Fifth and finally, we must recognize that dealing with the open-source challenge, even with the impressive progress of the past few years, will necessarily be a work in progress for some time to come. There is no one solution. Open source is not a traditional collection challenge. It will require multi-front strategies to master. It will take time for us to get this right.
Now, let me elaborate on these points. The world for the IC analyst has changed dramatically since the end of the Cold War. A decade ago, I thought of global coverage largely as the comprehensive strategy to collect against and analyze the Soviet Union – our single strategic threat. Today, global coverage for us means the responsibility to assess diverse, complex, and dispersed threats around the world, each on its own merits: terrorism; proliferation of WMD; organized crime; narcotics; and regional challenges from China and Russia, the Middle East, South Asia, and the Balkans. Significant advances in science and a substantial increase in knowledge-based technologies will add new issues and new levels of complexity to the challenges we face. The growing challenge of space and revolutionary progress in biotechnology, for example, presents threats as well as opportunities for the United States.
As I said earlier, the post-Cold War challenge has been increased by the revolution in information technology and telecommunications, which has fundamentally transformed the globe we cover, the service we provide consumers, and the workplace in which we function.
Information abounds. A growing volume of open-source material is relevant to our needs. Fifteen years ago, when I was working on the Balkans, information was scarce, foreign newspapers took weeks to get to my desk, and policymakers were willing to wait days or even weeks for me to produce a paper.
Today, everything moves faster. Everybody is better informed. Intelligence requirements, as a result, tend to be sharper and more time-sensitive. We receive newspapers and media reports often before the people in the countries themselves. Intelligence consumers will not tolerate waiting days for a response. They want it now. Technology makes us more efficient but this only increases the demand from our consumers.
As we look at the world we are charged to cover, governments are having less and less capacity to control information flows. International organized crime groups, terrorists, narcotraffickers, and proliferators, are taking advantage of such technology as well, bypassing governments, or seeking to undermine them when governments try to block their illegal activities.
In the years ahead, this will raise the profile of transnational issues that are already putting such heavy demands on intelligence collection and analysis.
In addition to traditional intelligence concerns—such as the future of Russia and China, political turmoil in Indonesia, and civil conflicts in Africa–– the new environment features many nontraditional missions such as peacekeeping, humanitarian assistance, sanctions-monitoring, information warfare, and combating international organized crime, as well as greater emphasis on such transnational issues as counterterrorism, counternarcotics, and counterproliferation and it contains the new challenges of cyberspace and real space, in threats to our space systems.
Many of these missions are operationally focused, requiring growing proportions of the analytic and collection work force to function in a crisis, ad hoc mode.
We are entering a world, moreover, in which we all will have access to a single network. We must still deal with terrorists, insurgents, and others who have hundreds of years of history fueling their causes – but chances are they will be using laptop computers, establishing their own websites, and using sophisticated encryption and weaponry their predecessors could not even have imagined.
At the same time, the information technology relationship between US government and industry has undergone a dramatic transformation. Today, government no longer dominates Research & Development and the information marketplace. Industry’s information technological R&D is focused primarily on commercial applications, and the Intelligence Community’s needs will increasingly be satisfied by the commercial sector.
The Information Glut
Of course, the key feature of the new information environment is simply that there is a lot more of it! During the Cold War, our job was piecing together bits of secret information. Each piece of raw intelligence was a carefully acquired golden nugget. Today, we’re still mining for information, but we’re facing an avalanche: not only from open sources, but also from classified collection systems. The situation reminds me of the time I was standing watch on a Navy ship and heard one officer ask another as he scanned the horizon with his binoculars: "My God, have you ever seen so much water?" And the second officer replied, "Yes, and we only see the top of it!"
Coping With the Information Glut
To serve our customers—who are smarter, more demanding, and want detailed answers in a heartbeat––analysts must first be trained to ask the right questions. It reminds me of another story I heard about a battlefield commander who wanted to know how the war was going. He called a lieutenant in from the front lines and said gruffly, "Lieutenant, in a word, how is the war going?" The timid lieutenant replied, "Good, sir." The commander wanted more information, so he said to the lieutenant, "Okay, in two words, how is the war going?" This time the lieutenant replied, "Not good, sir."
The point is when intelligence analysts query data bases, they need to know how to ask the questions in a way that will get useful answers. And we need analytical tools to help them extract the right data.
The number of sources and the overall amount of data to which an analyst has access makes the process of finding precise information, or hidden clues, very difficult. How can the analyst know where to start looking? What data might be relevant and what should be ignored? We believe there is a significant opportunity to use automated analysis tools—data mining and retrieval techniques—to help solve this problem.
Four years ago, CIA’s Directorate of Science and Technology and its Directorate of Intelligence collaborated in the formation of the joint Office of Advanced Analytic Tools (AAT), which works inside CIA with analysts to determine their needs and outside CIA with vendors to identify state-of-the-art collaborative and cognitive tools. Collaborative tools, as you know, facilitate quick and secure sharing of information worldwide. Cognitive tools help Agency officers to sift through large volumes of information—regardless of format or language—and to identify relationships and trends in the data.
Several such analytical tools are being developed to facilitate collaboration and assist in managing the information glut. Let me mention a few:
Clustering lets analysts exploit the most useful data sets first, helping us perform our warning function. Clustering is particularly helpful when the volume of information, as with open sources, makes it difficult to recognize meaningful patterns and relationships.
Link analysis can help establish relationships between a known problem and unknown actors, and can help detect patterns of activities that warrant particular attention.
Times series analysis can enable analysts to identify time trends, so that unusual patterns will be noticed.
Visualization allows analysts to see complex data in new and varied forms, including both link and time-series analysis.
Automated database population will allow analysts to be freed from the tedious and time-consuming function of maintaining databases. This will reduce the potential for errors and inconsistencies.
Many of these tools and techniques are available to the private sector, as well as government. They enhance our ability to filter, search, and prioritize potentially overwhelming volumes of information.
One of the strongest and most consistent needs of our analysts is to search and exploit both classified and unclassified information from a single work station. The Community is also working on ways to standardize information and tag it using metadata—or reference information––so it will be easier to search, structure, and populate into databases.
FBIS is developing a single, open source "portal" that will organize and cross-reference information that FBIS has collected via the Internet, as well as FBIS products and other multimedia material.
The "portal"—which is expected to be fully operational by 2002—will provide analysts with a one-stop shop for all open-source intelligence, whether collected by FBIS or not.
Material on the portal will be indexed, archived, and accessible via a powerful search engine that is easy to use. And analysts will be able to access this portal from their desktops.
Today’s intelligence analysts are as comfortable in cyberspace as in the office space of top consumers. The Washington-based analyst today can send a message and get a response from a remote country post faster than it used to take to exchange notes by pneumatic tube with counterparts in the same building.
We see significant opportunity in the area of collaborative tools. As you know, the Community consists of 13 agencies, several DCI centers, the National Intelligence Council, and literally hundreds of collection and analysis offices. The problem of sharing data among such a large number of organizations is immense.
For one thing, different agencies have different security standards. All organizations have private intelligence holdings that are extraordinarily sensitive. We have to resolve the issue of multilevel security and need-to-know concerns through the development of robust and flexible communities of interest.
As I see it, two types of collaborative tools are needed: collaboration in the production process––to increase speed and accuracy––and expertise-based collaboration––to enable a team of analysts to work on a project for several weeks or months. Several collaborative tools currently available or soon to be deployed include the capability to share both textual and graphical information in real time. Future requirements will be for broad deployment of collaborative tools, relying on mature commercial off-the-shelf platforms performing to standards that allow interoperability across the Intelligence Community.
These new collaborative tools will allow analysts to discuss contentious analytical issues, share information like maps, imagery, and database information, and coordinate draft assessments, all on-line, from their own workspaces, resulting in substantial savings of time and effort.
Another important area is distributed knowledge. The Intelligence Community will never have a database that contains all information that is available to all organizations, due to the individual missions of each organization. But the ability to share major holdings and to present an integrated view to the analyst’s desktop is critical—and is no easy task!
Finally, we have some challenges that few private sector organizations face. We deal in foreign languages extensively: lots of them. FBIS, for example, translates and disseminates information in many different languages. A number of the tools that exist today, though getting better, still do not function well in such an environment.
Working With the Private Sector
The Intelligence Community recognizes that partnerships with outside academic and technical experts, as well as vendors, are essential to enabling us to stay on top of the information technology curve. It has been said that, "Opportunities are like sunrises. If you wait too long, you miss them." We don’t intend to miss today’s opportunities because we’re too inwardly focused. In short, we need outside help, and we know it!
In 50 years, we’ve gone from large, stationary mainframes with a handful of dumb work stations to portable multi-service devices that will communicate, compute, and run your office. This represents a dramatic leveling of the costs of information and affects the way the Intelligence Community does its work. But in many ways, the Community still thinks and organizes itself with immobile information systems.
In fact, we’re investing even greater amounts in stationary hardware systems. But many of our targets—terrorists, narcotraffickers, and organized crime syndicates—are becoming increasingly mobile in their operations. Perhaps someday private industry can come up with ways to liberate us –I hope soon—from our information cubicles, while at the same time ensuring the security of our work.
Looking at more recent developments, I am pleased to report that CIA has launched a new nonprofit corporation designed to bring the best of the academic, business, and private research worlds together to exploit new and emerging information technologies.
The new corporation was first launched in February 1999 as In-Q-It but had its name changed to In-Q-Tel in December 1999 to prevent confusion between its organization and the financial software giant Intuit. "In-Tel" is self-explanatory, while the "Q" stands for technical innovation (derived from the James Bond character who developed Bond’s spy gear). In-Q-Tel is a collaborative venture among the government, industry, and academia, whose mission is two fold:
First, to accept strategic problems and develop a "portfolio" of innovative and unconventional information technology solutions, ranging from exploration to demonstration; and
Second, to fuel private research, development, and application of information technologies of strategic national interest for the benefit of all partners.
In-Q-Tel will not conduct the research itself; rather, it will orchestrate the work of numerous partner organizations working in teams. In-Q-Tel’s initial projects focus on four interrelated intelligence challenges:
First, Agency use of the Internet—particularly Internet search and privacy issues.
Information security is a cross-cutting issue that permeates all organizational functions. As such, In-Q-Tel will engage information security from the following perspectives: hardening and intrusion detection, monitoring and profiling of information use and misuse, and network and data protection.
Third, In-Q-Tel will focus on a few distinct types of analytic data processing capabilities: geospatial and multimedia data fusion/integration, all source analysis, and computer data forensics.
Fourth, In-Q-Tel is tasked with dealing with the problems associated with the Agency distributed information technology infrastructure, which is both organizationally segmented and geographically dispersed.
Returning to the imperative to develop a renewed US Intelligence Community, let me close by emphasizing that the Community can succeed only if it exploits the changes taking place in the larger information industry—and we know this.
We will always have security concerns, and we must address them. But we cannot allow those concerns to deter us from acknowledging the opportunities inherent in the emerging environment. We are, at the end of the day, an information-based business that will spend more and more time chasing bad guys with access to information-based technologies.
As the heirs to the men and women who solved the secrets of the Enigma and who invented satellite reconnaissance, we are prepared to face the technological future with confidence.
For most of our history, the Intelligence Community has operated as an industrial enterprise, with compartmentation as a key operating metaphor. We have created a set of impressive but inflexible organizations. Now we must adjust to a world in which networks defeat hierarchies, and in which agility is a prerequisite for organizational success.
We in the Intelligence Community have committed ourselves to a corporate strategy that will leverage the best practices and resources of the whole government and the private sector to provide the President and US policymakers the information advantage they need.
Let me stop there and take your questions.