FAS

Basic Failures Abound in Classification Program

01.14.09 | 2 min read | Text by Steven Aftergood

“At a time where we would expect to find increasing stability in the [national security classification] program, we are instead finding failure with the implementation of basic requirements,” wrote William J. Bosanko, director of the Information Security Oversight Office (ISOO), in the latest ISOO annual report to the President (pdf).

Out of more than 1,000 classified documents examined by ISOO last year, “the appropriateness of classification was subject to question in over 25 percent,” Mr. Bosanko reported.  See the FY 2008 ISOO Report to the President, transmitted January 12, 2009.

In what may be the report’s most significant finding, ISOO discovered that the majority of classification guides used by government agencies to prescribe exactly what information should be classified at what level are badly out of date.

“Overall, 67 percent of the guides agencies reported as being currently in use had not been updated within the past five years,” the ISOO report said.  In effect, agencies are continuing to impose outdated classification restrictions on newly generated information.

This finding underscores the utility of, and the need for, an agency-by-agency “scrub” of all classification guides in order to eliminate obsolete classification practices.  (For more on this approach, see “Overcoming Overclassification,” Secrecy News, September 16, 2008.)

The new ISOO report also had some favorable news.  The number of original classification authorities (who are authorized to designate new information as classified) declined slightly.  The number of original classification decisions — new secrets — dropped by 13 percent.  For the fourth year in a row, a majority of new classification actions were assigned a declassification date of ten years or less.  The number of classification challenges within the executive branch disputing the classification status of certain information rose to 436 formal challenges from 275 the year before.  The ISCAP, which reviews appeals of declassification requests that have been denied, declassified a greater percentage of information than in past years.

But in general, declassification languished.  “The overall number of pages reviewed and pages declassified by Executive branch agencies has declined significantly from previous years.”  And it is unlikely that agencies will meet a December 31, 2009 deadline for automatic declassification of 25 year old records that contain multiple agencies equities (or interests), the ISOO report said.

In the end, the classification system can only work as well as government officials want it to work, the ISOO report concluded.

“Ultimately, the success or failure [of agency classification policies] depends on the commitment of the agency heads and senior agency officials to the classified national security information program established by the President,” ISOO said.

publications
See all publications
Clean Energy
Blog
Fixing a Broken Market: A Plan for Cheaper Freight, Cleaner Air, and American Truck Leadership

Americans are paying too much for almost everything, because the United States has long treated its trucking industry as an artifact to be preserved rather than as an opportunity for innovation.

06.16.26 | 9 min read
read more
Emerging Technology
Report
SOURCE CODE: A Policy Agenda for Fostering Trust and Fairness in AI

These ideas aim to advance the detailed policy solutions needed to foster public trust and implement fairness in the adoption of AI across diverse domains, from healthcare and government benefits to rural access, education, and worker protections.

06.11.26 | 17 min read
read more
Emerging Technology
day one project
Policy Memo
Move Algorithmic-Driven Pay and Scheduling Systems From Surveillance Pay to Fair Wages

The evidence is clear: algorithmic pay-setting is established in app-based work, and payroll/timekeeping failures show how software can produce systemic wage harm at scale

06.11.26 | 15 min read
read more
Emerging Technology
day one project
Policy Memo
How State Leaders Can Put People First in AI Decision-Making

While a few states have taken steps to implement decision-making mechanisms for certain AI systems, too many leaders are simply accepting narratives about AI’s purported public benefit at face value – jumping to the “how” of AI implementation before thoroughly vetting potential systems and deciding whether they are appropriate to use at all.

06.11.26 | 17 min read
read more