It looks like no one in that thread could dig up any experimental evidence for it. Manufacturer recommendations are not to be trusted. This looks like a job f

It looks like no one in that thread could dig up any experimental evidence for it. Manufacturer recommendations are not to be trusted. This looks like a job for some identical knives, some knife-use simulator robots, several dishwashers with a standardized typical load, hand-washing sinks, a pile of dishwasher detergent, a pile of hand-wash detergent, and an objective measurement device for knife sharpness. My essential laziness (and lack of research grant money) prevents me from conducting the experiment myself, however. I'll just take my chances with the dishwasher.