New Executive Order Aims to Avoid Declass Deadline
Development of a new executive order on classification of national security information is now proceeding at an accelerated pace in order to preempt a deadline that would require the declassification of millions of pages of historical records next month.
A revised draft executive order was circulated to executive branch agencies by the Office of Management and Budget on November 16, with agency comments due back today, November 23. A final order is likely to be issued by the end of this year.
There is an incentive to complete the development of the executive order before December 31, 2009 because of a deadline for declassification of historical records that falls on that date. Under the current Bush executive order, classified records that are at least 25 years old and that have been referred from one agency to another because they involve multiple agency interests are supposed to be automatically declassified at the end of this year. (See E.O. 13292, section 3.3(e)(3)).
But in order to meet this December deadline, several agencies would have to forgo a review of the affected historical records, which they are unwilling to do. And so it seems they will simply be excused from compliance. But in order to modify the deadline in the Bush order, it will be necessary to issue another executive order. If the comprehensive new Obama order on classification policy (which would assign processing of such records to a National Declassification Center that does not yet exist) is not ready for release by December 31, then another stand-alone order would have to be issued, canceling or extending the looming deadline. And officials are reluctant to issue such an order since they say it would be awkward for the avowedly pro-openness Obama Administration to relax or annul a declassification requirement that was imposed by the ultra-secret Bush Administration.
In fact, the whole process has become an awkward mix of exaggerated and deflated expectations. The failure of the Bush Administration’s declassification deadline to take hold this year does not augur well for new, more ambitious efforts to advance classification reform. If the “automatic declassification” procedures that were prescribed in prior executive orders are not “automatic” after all, and if binding deadlines can be extended more or less at will, then any new declassification requirements in the Obama order will be similarly subject to doubt or defiance.
The latest draft executive order has not yet become publicly available, though officials said they expected it to leak, as did a previous draft dated August 4. “It includes some notable differences” from the earlier draft, said one official. But another official said “It’s basically the same as the draft you already have.” (See “Draft Order Would Set New Limits on Classification,” Secrecy News, September 29, 2009.)
Ironically, today’s classification system seems to function more effectively in preventing public access to aging archival records than it does with respect to certain present-day information.
Thus, Defense Secretary Robert M. Gates told reporters on November 12, “I have been appalled by the amount of leaking that has been going on” about classified Administration deliberations on Afghanistan policy and other matters.
But from a different point of view, others may be appalled that Secretary Gates’ own Department still retains classification restrictions on historical records dating back to the Korean War, and even from World War II, and that it otherwise resists modernization and correction of the cold war classification system.
Some general background on the national security classification system from the Congressional Research Service can be found in “Security Classification Policy and Procedure: E.O. 12958, as Amended” (pdf), November 3, 2009.
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.
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
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.
When properly structured — with specific numeric targets, secured financial obligations, independent monitoring, and meaningful enforcement — CBAs transform data center deals into durable community partnerships.