A new study published by the CIA Center for the Study of Intelligence calls for a fundamental reconceptualization of the process of intelligence analysis in order to overcome the “pathologies” that have rendered it increasingly dysfunctional.
“Curing Analytic Pathologies” (pdf) by Jeffrey R. Cooper has been available up to now in limited circulation in hard copy only. Like several other recent studies critical of U.S. intelligence, it was withheld from the CIA web site. It has now been published on the Federation of American Scientists web site.
Author Cooper provides a thoughtful critique that notes the intrinsic difficulties of intelligence analysis and observes how current organizational practices have exacerbated them.
“The Intelligence Community presently lacks many of the scientific community’s self-correcting features,” he writes.
One major impediment to improving analysis is the hypertrophied secrecy practices that prevail in intelligence organizations.
“Unfortunately, the more that evidence and judgments are restricted in dissemination by compartmentation and distribution limitations, the more likely it is that questionable judgments will pass unchallenged.”
Fundamentally, the whole concept of the “intelligence cycle” — referring to the conventional sequence of collection, processing, analysis and dissemination — is misleading, Cooper argues, and should be jettisoned.
“With its industrial age antecedents, it usually conveys the notion of a self-contained ‘batch’ process rather than a continuous spiral of interactions.”
Taking Cooper’s thesis seriously, one could respectfully say that his new study embodies some of the defects in intelligence analysis that he writes about.
Thus, the study presents what is essentially one moment in an ongoing conversation and freezes it in a nicely produced but static document. This reflects the kind of spurious finality that Cooper dismisses as the “conceit of finished intelligence.”
And then the CIA, by disseminating the document in hardcopy only, sharply limited its audience and effectively precluded a “continuous spiral of interactions” regarding its contents.
That last part, at least, can be corrected.
See “Curing Analytic Pathologies: Pathways to Improved Intelligence Analysis” by Jeffrey R. Cooper, Center for the Study of Intelligence, December 2005 (5 MB PDF).
Author Cooper said he would welcome feedback from interested readers. Comments can be posted here on the Secrecy News blog. Alternatively, Cooper’s contact information can be obtained from Secrecy News.
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.