WWII Atomic Bomb Project Had More Than 1,500 “Leaks”
The Manhattan Project to develop the first atomic bomb during World War II was among the most highly classified and tightly secured programs ever undertaken by the U.S. government. Nevertheless, it generated more than 1,500 leak investigations involving unauthorized disclosures of classified Project information.
That remarkable fact is noted in the latest declassified volume of the official Manhattan District History (Volume 14, Intelligence & Security) that was approved for release and posted online by the Department of Energy last month.
In several respects, the Manhattan Project established the template for secret government programs during the Cold War (and after). It pioneered or refined the practices of compartmentalization of information, “black” budgets, cover and deception to conceal secret facilities, minimal notification to Congress, and more.
But wherever there are national security secrets, it seems that leaks and spies are not far behind.
During the course of the Manhattan Project, counterintelligence agents “handled more than 1,000 general subversive investigations, over 1,500 cases in which classified project information was transmitted to unauthorized persons, approximately 100 suspected espionage cases, and approximately 200 suspected sabotage cases,” according to the newly declassified history (at pp. S2-3).
Most of the 1,500 leak cases seem to have been inadvertent disclosures rather than deliberate releases to the news media of the contemporary sort. But they were diligently investigated nonetheless. “Complete security of information could be achieved only by following all leaks to their source.”
In 1943, there were several seemingly unrelated cases of Protestant clergymen in the South preaching sermons that alarmingly cited “the devastating energy contained in minute quantities of Uranium 235” (while contrasting it with “the power of God [that] was infinitely greater”). The sermons were eventually traced back to a pamphlet distributed by a Bible college in Chicago, which was determined to be harmless. Other disclosures cited in the history involved more serious indiscretions that drew punitive action.
“Since September 1943, investigations were conducted of more than 1500 ‘loose talk’ or leakage of information cases and corrective action was taken in more than 1200 violations of procedures for handling classified material,” the history said (p. 6.5).
“Upon discovery of the source of a violation of regulations for safeguarding military information, the violator, if a project employee, was usually reprimanded, informed of the possible application of the Espionage Act, and warned not to repeat the violation.”
Fundamentally, however, information security was not to be achieved by the force of law or the threat of punishment. Rather, it was rooted in shared values and common commitments, the Project history said.
“Grounds for protecting information were largely patriotism, loyalty to the fighting men, and the reasoning that the less publicity given the Project, the more difficult it would be for the enemy to acquire information about it and also, the greater would be the element of surprise” (p. 6.13).
The only other remaining portion of the official history, Foreign Intelligence Supplement No. 1 to Manhattan District History Volume 14, was also published online last month. It provided an account of U.S. wartime intelligence collection aimed at enemy scientific research and development. Some information in that volume was deleted by the Central Intelligence Agency.
The entire thirty-six volume Manhattan District history has now been declassified and posted online.
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.
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.