INTELLIGENCE COLLECTION, ANALYSIS, AND DISSEMINATION

`Hard Target' Human Intelligence

The invaluable contributions of accurate Human Intelligence (HUMINT) to United States efforts in Operation Enduring Freedom, Operation Iraqi Freedom, and the global war on terrorism are evident.

Particularly in the context of the 107th Congress's Joint Inquiry into the terrorist attacks of September 11, 2001, various Committee Members expressed concern about the need for more vigorous HUMINT collection--especially unilateral collection--under non-official cover and from non-traditional HUMINT `platforms.' Some experts have even suggested the need for the creation of a small, highly-specialized semi- or fully-independent HUMINT entity charged with collecting against non-traditional targets and rogue states that traditionally have proven highly resistant to HUMINT penetration involving traditional official-cover operations.

Without endorsing such a radical solution at this time, the Committee attaches the highest degree of importance to far more aggressive and sustained non-traditional HUMINT collection program. The Intelligence Community must act now to meet the United States requirement for much improved HUMINT collection against hard targets. This will require diligent effort and new approaches to HUMINT management within existing agency components. The Committee hopes and expects that the Director of Central Intelligence will ensure the implementation and success of such changes within the Intelligence Community.

Pilot program on analysis of signals and other intelligence by intelligence analysts of various elements of the Intelligence Community

The Committee has become increasingly concerned in recent years about bureaucratic and cultural obstacles to effective information and data sharing. Such resistance to data access by `outsiders' within the Intelligence Community--let alone to other entities such as analysts at the new Department of Homeland Security--causes at least three serious problems.

First, it impedes the ability of the Intelligence Community to adopt state-of-the-art data-mining and analytical tools that are badly needed to help analysts cope with the flood of information brought in by collection components. Cutting-edge analytical tools, many of which are already in use in the private sector, increasingly involve innovative automated or computer-assisted tools to perform large-scale, multi-database analysis and pattern recognition. Using such approaches within the Intelligence Community, however, cannot proceed far without a significant revision of current orthodoxy as to information `ownership' and control.

Second, barriers to data access inhibit the Intelligence Community's ability to understand, correlate, and assess information that they already possess. Data-control restrictions sometimes impede sharing within an individual element of the Community, as well as between elements, limiting the effectiveness of analytical work far beyond what is necessary to protect highly sensitive information from undue risk of compromise.

Third, barriers to data access prevent the Community from employing other elements' analysts in understanding available information--both for the basic purpose of reducing data overload, and for more sophisticated goals like applying fresh analytical perspectives and experience to existing analytical tasks. The Committee supports additional analytic views to issues, and those views can only be enriched when informed by access to all available information and data.

The NSA, in particular, is an analytical organization that is far too small to handle the volumes of data that it collects. The reluctance of NSA to give other agency analysts access to data that NSA analysts do not have time or priority to review--even when such analysts are as well trained as NSA personnel in protecting `U.S. person' information--has prevented the use of non-NSA analytic manpower to help narrow the gap between collection and analysis and to ensure that more of the unevaluated NSA data is reviewed by an analyst. It has also drained NSA of trained analysts, because the agency elects to send many of its analysts to other agencies and organizations to supervise and regulate the small degree of NSA information sharing that does occur.

If other agency analysts were properly trained in the rules and procedures governing the handling of SIGINT information (as many are) and these analysts enjoyed the trust of NSA, such analysts would provide `value added' beyond their numbers. Many NSA liaison officers who are now situated in other Intelligence Community agencies to provide those agencies

with the ability to access NSA data could return to full-time NSA analytical work.

It has proven difficult to achieve significant improvements in data sharing and information access within and between elements of the United States Intelligence Community. The events of September 11, 2001--and the record of the Community's preparedness for these terrorist atrocities, as detailed by the Senate and House Intelligence Committees' Joint Inquiry during the 107th Congress--make immediate action and cooperation imperative. Accordingly, in Section 314, the Committee requires that concrete steps be taken to pave the way for a future of vastly improved data sharing and information analysis within the Intelligence Community.

Section 314 requires NSA to develop a pilot program to improve the ability of analysts in other Intelligence Community elements to obtain access to and analyze data collected and held by NSA, while retaining appropriate handling safeguards. The pilot program's objectives are: (a) to augment the Intelligence Community's ability to undertake true `all-source fusion' analysis in support of intelligence requirements by helping build a legal and practical foundation for increased inter-agency cooperation and data sharing; (b) to increase to the maximum practicable extent the proportion of NSA-collected information that is reviewed and assessed by intelligence analysts; and (c) to reduce the drain on NSA analytical manpower caused by current barriers to inter-agency information sharing.

Not later than December 31, 2003, therefore, NSA must begin to implement a program to:

a. Develop efficient and effective methods for certifying that designated analysts from other agencies are properly trained in the relevant procedures for handling SIGINT information and for `minimizing' any `U.S. person' information that might be contained therein so that such analysts may be given access to NSA databases in an identical fashion to NSA analysts; those analysts from other agencies will be designated as requiring such access by the head of their parent agency, who will retain full accountability for the analytic products produced by such agency's analysts; and

b. Explore and improve innovative ways to allow other agencies to apply their analytical expertise to NSA data, including the use of `detailees in place' (i.e., other-agency employees who are notionally detailed to NSA, thus becoming part of the SIGINT enterprise while remaining at their home agency).

Pilot program on training for intelligence analysts

Current programs that encourage students to pursue educational programs relevant to national security or foreign language training have not produced the number of qualified analysts or foreign language experts necessary to meet the ongoing needs of the Intelligence Community. Although the David L. Boren National Security Education Act of 1991, Title VIII, Public Law 102-183 (Dec. 4, 1991) moves in the right direction, the David L. Boren National Security Education Program (NSEP) places students in national security positions throughout the Federal Government, not merely within the analytic components of the Intelligence Community.

To address the shortage in language proficient and area expert analytic capabilities within the Intelligence Community, the Committee proposes the Director of Central Intelligence establish a ROTC-like Intelligence Analyst Program. This program should seek to increase the number of qualified entry-level intelligence professional analysts available to the Intelligence Community. The goal of the program should be to recruit entry-level analysts and operations specialists with enhanced analytic and foreign language skills who are committed to a career in the Community. The Committee believes this program should be national in scope (conducted at universities throughout the United States), able to identify individuals interested in working in the Intelligence Community, able to provide financial assistance to participants, and capable of providing guidance to participants in selecting courses that would be most useful for an intelligence analyst's career. The program should also include educating participants on the various analytic specialties and opportunities within the Intelligence Community. Prerequisites of the program, and financial assistance thereunder, should include the ability to obtain a security clearance and a commitment for service within the Intelligence Community.

The Committee is pleased that the ADCI/A&P has expressed strong support of the goals of this initiative. Moreover, the Committee believes that the ADCI/A&P is the proper entity to manage such a pilot program for

approximately 150 students in fiscal year 2004. Therefore, the Committee recommends an increase of $8.0 million to the ADCI/A&P to create and manage this pilot program.

Report on modifications of policy and law on classified information to facilitate sharing of information for national security purposes

The Committee is concerned that Executive Orders 12333 and 12958 and related regulations and policies may inappropriately limit the effective sharing of intelligence information and data. The Committee therefore has several provisions within the Bill to improve sharing within the Intelligence Community to enhance the quality and timeliness of intelligence products. Furthermore, it is the sense of the Committee that information sharing will become increasingly important as the Department of Homeland Security endeavors to move critical information in both directions between the Intelligence Community and regional, state, and local governments.

Of particular concern to the Committee are various sections in Executive Orders 12333 and 12958 that the Administration should expeditiously review to accurately reflect the movement to electronic data collection and storage and to address the requirement to more effectively assess and share pertinent national security information. The war on terrorism and the proliferation of weapons of mass destruction require foreign and domestic and national and local partners to effectively collaborate on analysis and coordinate on operations. Achieving this aim will require the U.S. Government to move beyond the paradigm imposed by individual agency `ownership' of information. In that regard, the Committee notes that agencies that collect information often do not have the requisite analytic workforce to fully exploit the data they collect. Further, expanded access to data--providing it is done securely--will foster more rapid access and more competitive analysis and exploitation. Notwithstanding extremely sensitive security, operational, and related matters, the President should continue to encourage broader and more secure exchange of information within the Intelligence Community and between the Community and its many consumers. Executive Order 12958 should be revised not only to remove restrictive impediments within the Executive Order as to the inter-agency sharing of classified information, but also to facilitate sharing and access (except in narrowly defined circumstances). Executive Order 12333 similarly needs to be revised such that other organizations besides NSA can engage in SIGINT activities, specifically analysis of SIGINT information that has been lawfully collected.

Revisions to Executive Orders 12333 and 12958 are needed, but represent one of many issues that need to be addressed to achieve greater teamwork in the defense of the nation. Technical solutions, such as the need to implement machine-enabled processes to automatically tag data, are needed to facilitate efficient access and analysis. Further, the Intelligence Community must recognize that information sharing cannot succeed without revised security policies and technologies. This Bill, therefore, requires several related reports, including reviews of security clearance procedures, the threats to networks posed by `cleared insiders,' and the growing reliance of the United States on foreign hardware and software. Only with a broad approach, encompassing policy and technology and security and sharing, can we achieve the maximum advantages offered by modern information technologies and a highly-trained and motivated workforce.

The Committee requests that the President review Executive Orders 12333 and 12598 and submit a report within 6 months on potential changes to the Executive Orders or legislative actions which could be applied to facilitate information sharing and data access across the Intelligence Community.

Report on data-mining capabilities for the Intelligence Community

Data mining is emerging as potentially one of the most valuable tools for Intelligence Community analysts. Data mining involves the use of data analysis tools to discover previously unknown, valid patterns and relationships in large data sets. This technology has the potential to provide intelligence analysts with the capability to identify terrorists, to recognize the development and proliferation of weapons of mass destruction, and to detect illicit narcotics activity (e.g., communications, money transfers, and travel) by examining voluminous records.

The Committee is concerned, however, that components of the Intelligence Community are investing in a variety of data-mining capabilities without sufficient coordination of Community-wide data-mining requirements. Strategic planning in this area is vital to ensure that there is no redundancy of effort and that interface standards are in place to enable collaboration and cross-program application, as required.

Accordingly, the Committee directs that the Chief Information Officer for the Intelligence Community (CIO) and the ADCI/A&P jointly review data-mining capabilities throughout the Intelligence Community and assess which capabilities meet Intelligence Community analytic requirements. The CIO will publish guidelines to the Community on standards and protocols to enable cross-agency interfaces for migration and data-level information exchange. The results of this assessment should be included in a written report to the House and Senate Intelligence Committees (submitted no later than December 1, 2003) and should include funding requirements for respective data-mining capabilities.