Of course, not all inmates are psychopathic, but 15-25% manifest those traits, according to https://www.apa.org/monitor/2022/03/ce-corner-psychopathy.
It's not clear that everyone who displays some of those traits is going to display all the others, that a pattern of risky violence means someone is automatically beyond help, that the stats on psychopaths aren't including lots of people with other underlying problems that fully explain their behaviour, or that anyone else who goes to a therapist is doing anything radically different from simulating normal human behaviour.
the reference I gave is not pop science.
Your debunking and pooh-pooh'ing is unpersuasive. This is a professional association publishing that, and they do have shrinks in prison.
"About 1.2% of U.S. adult men and 0.3% to 0.7% of U.S. adult women are considered to have clinically significant levels of psychopathic traits. Those numbers rise exponentially in prison, where 15% to 25% of inmates show these characteristics (Burton, B., & Saleh, F. M., Psychiatric Times, Vol. 37, No. 10, 2020). "
Your reference has no bearing on the answer to that question and in fact notes that 30% of the population display some traits of psychopathy, which is the entire point here. If you squint hard enough, you can label 30% of the population as psychopaths; if you're also convinced it's useless to send them to therapy, you can use the label to deny them mental health care and stick them in prison.
I'm worried that a broad and underspecified definition of psychopathy combines with an overstated inability to treat them, in a way that is morally bankrupt and tantamount to just giving up.
The weakest in society are those who are taken advantage of by the morally bankrupt, not the morally bankrupt themselves.
However, I agree that what we are currently doing is worse than almost every alternative.