hn.today

The secret Cold War bioweapons test that exposed millions in the Bay Area

sfgate.com3 points0 comments
Screenshot of The secret Cold War bioweapons test that exposed millions in the Bay Area

In September 1950, U.S. Army scientists secretly sprayed the San Francisco Bay Area from a converted minesweeper with Serratia marcescens to model how a biological agent might disperse. The aerosolized bacteria reached much of the city and surrounding communities; Army monitoring reported residents inhaling thousands of particles per minute. Within days Stanford Hospital recorded an unprecedented cluster of Serratia infections, including a 75-year-old retired PG&E employee, Edward Nevin, whose autopsy showed fatal Serratia endocarditis. The tests continued for years - other vulnerability experiments included releasing bacteria in subway tunnels - and remained classified until parts were declassified in 1976, when Nevin’s son discovered the operation and sued the government. Judges ultimately dismissed the claim; several program leaders publicly defended the experiments.

The exposure left lingering public-health and ethical consequences. Medical clusters of Serratia infections and later incidents - most notably a 2001 outbreak tied to contaminated cortisone shots - prompted questions about whether the 1950 release permanently altered the Bay Area’s microbial ecology. Scientists say comparison with the Army’s original strain would be decisive, but those samples and some documents have never been produced. The episode is remembered as a Cold War policy that prioritized perceived national security over civilian safety and informed ongoing debates about government secrecy, accountability and the risks of human-subject biodefense testing.

Read on sfgate.com0 comments on Hacker News

Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.

More in Security

Can gzip be a language model?

Can gzip be a language model?

Gzip, a standard compression tool, can be viewed as a form of language model because it predicts text continuations based on compression patterns. By priming gzip with a corpus and measuring compressed lengths, it can generate text that reflects learned structure without neural networks. (nathan.rs)

The daily digest

Today's best Hacker News stories, summarized and screenshotted, one email a day.