In the 1960s, the Central Intelligence Agency spent five years and millions of dollars surgically implanting a listening device inside a live cat. The project was called Acoustic Kitty.

The idea came from the CIA's Directorate of Science and Technology, the division responsible for the agency's most ambitious and frequently absurd technical programmes. The Cold War was at its peak. The Soviets were everywhere. The Americans needed new ways to listen, and someone in a meeting room in Langley suggested: what if the microphone was alive?

The veterinary surgeons went to work. They implanted a small microphone in the cat's ear canal. They ran a thin wire antenna through the cat's fur from the base of its skull to the tip of its tail. They embedded a radio transmitter and a battery in the cat's abdomen. The surgery took roughly an hour. The cat survived. It could walk, eat, and behave like a normal cat, while transmitting every sound within earshot to a CIA listening post nearby.

There was one problem. It was a cat.

Training a cat to walk to a specific location, sit near a specific person, and stay there long enough to record a useful conversation turned out to be, in the words of a declassified CIA memo, not practical. The cat wandered. It got hungry. It got distracted. It did what cats do, which is whatever they want.

The project reportedly cost between fifteen and twenty million dollars. According to Victor Marchetti, a former CIA officer, the cat was deployed for its first field test near a park in Washington, directed to walk toward two men sitting outside the Soviet compound. It crossed the street, was hit by a taxi, and that was the end of Acoustic Kitty.

The CIA's own account is gentler. Robert Wallace, a former director of the Office of Technical Service, later said the cat was not killed by a taxi. The programme was simply shelved because the cat could not be reliably directed. The equipment was surgically removed, and the cat, he claimed, lived a long and happy life.

Either way, the conclusion was the same. The project was cancelled. The final memo, declassified in 2001, praised the work as a "remarkable scientific achievement" but noted it had no practical application.

The CIA spent millions to prove that you cannot tell a cat what to do. Anyone who has owned a cat could have told them that for free.

I do one of these a day. Dead serious. Subscribe.