while it's technically possible to hear above 20 kHz, outside of a laboratory it's far more likely that you're hearing intermodulation distortion, or simply background noise: at least on Linux, the audio output can be disabled completely when not in use. therefore, there is background noise only when the audio output is activated.
Are you asserting that GP is confused about what they are hearing, or are you saying that the ultrasound is creating some kind of unpleasant interference that is being perceived instead?
I've been able to hear in unusually high ranges my whole life into my mid-30s. Office, home, urban environments... distance measuring equipment, devices intended to repel birds and mosquitoes... things that are loud but "too high to hear" that wind up for me being "yes, that is indeed very loud and very high".
Being relatively young and having a childhood history of asthma could get them hearing a little higher than 20 khZ as well, so it's not out of the question that they are hearing the sound itself.
But the most likely thing is that they are hearing some sort of undertone or artifact created in the process of the signal generation.
