It is possible that the "population count" instruction has been included in the instruction sets of most American supercomputers at the request of NSA, which was an important customer for them.
Nevertheless, the first computer having this instruction was a British computer, the Ferranti Mark I (February 1951).
The name used by Ferranti Mark I for this instruction was "sideways add".
Also notable was that Ferranti Mark I had the equivalent of LZCNT (count leading zeroes) too.
Both instructions are very useful and they are standard now for modern instruction sets, but they were omitted in most computers after Ferranti Mark I, except in expensive supercomputers.
Moreover, Ferranti Mark I included a hardware random number generator, another feature useful for cryptography, which was reintroduced only recently in modern CPUs.
Hardware random number generators do have some security issues though. Linux devs were opposed to solely relying on them, because they can be compromised by the vendor [1]. So they are at best used in algorithms that they can not compromise (still in [1], but lower, in the comments).
The security issues are not with hardware random number generators in general, but with those that are included inside complex devices like monolithic CPUs or TPMs, so that the owners of those devices cannot verify that the RNG's really do what they are claimed to do.
Discrete hardware RNG's, like that of the Ferranti Mark I, are perfectly secure.
For a modern device, the best way to implement a hardware RNG is to just include an ADC (analog-digital converter) input. Then you may connect externally on the PCB some noisy analog amplifier, e.g. one which has a noisy resistor or diode at its input. Digitizing the noise with the ADC will provide the random numbers and the ADC input can be isolated and tested separately at any time, so the user can verify that there is no hidden functionality.
Most microcontrollers have ADC inputs, so it is easy to add a secure hardware RNG for them. The same could be done for a personal computer by making a noisy amplifier that can be plugged in the microphone input, or by making a USB device with a microcontroller.
I remembered something, and I want to say as an aside, for anybody reading that at one point has to design a toy RNG from an ADC, as I had to some years ago, you should not take the last bits as they are--as was my first thought--, you should pass them through something like the von Neumann corrector [1].
Indeed, AMD has more than once shipped CPUs in which the random-number instruction would always yield the same value, that had to be monkey-patched to yield apparently random numbers. A valuable hint.
Nevertheless, the first computer having this instruction was a British computer, the Ferranti Mark I (February 1951).
The name used by Ferranti Mark I for this instruction was "sideways add".
Also notable was that Ferranti Mark I had the equivalent of LZCNT (count leading zeroes) too.
Both instructions are very useful and they are standard now for modern instruction sets, but they were omitted in most computers after Ferranti Mark I, except in expensive supercomputers.