The AIR (American Institutes for Research) review (September 1995)

In the summer of 1995, the U.S. Congress directed the CIA to evaluate its own remote-viewing program. The CIA contracted the work out to the American Institutes for Research, a Washington research nonprofit. AIR assigned the technical review to two researchers with sharply different priors: Jessica Utts, a statistician at the University of California, Davis, and Ray Hyman, a psychologist at the University of Oregon and a longtime critic of parapsychology.

Utts and Hyman were given access to the SRI and Science Applications International Corporation data archives. They examined the laboratory experiments that had been used to argue the phenomenon was real, the operational sessions that had been tasked by intelligence agencies, and the program management records. They published independent reports as appendices to the final AIR document. The two researchers agreed on the underlying statistics. They disagreed sharply on what the statistics meant.

Utts concluded that the laboratory evidence for remote viewing was statistically significant beyond reasonable doubt, comparable to effects accepted in other areas of psychology, and that the question of whether the effect was real should be considered settled in favor of further research. Hyman conceded that the statistics were unusual and that the simplest explanations (cueing, fraud, selection bias) had been ruled out in the better experiments, but argued that the absence of a known mechanism and the failure of replication outside SRI required treating the finding as unproven.

On the operational question, both reviewers agreed: there was no evidence that the program had produced intelligence useful enough to justify its continued government funding. The director of the AIR review, Michael Mumford, recommended termination. The CIA accepted the recommendation. The program ended in November 1995.

The full AIR report, including the Utts and Hyman appendices, was declassified along with the rest of the Stargate archive in the 2017 release. It runs to several hundred pages. It is the single most cited document in the academic literature on remote viewing, both by defenders of the program and by its critics, who read the same pages and reach opposite conclusions.