I tried to propose that I would use the GPU cluster to improve medical diagnostics, but it insisted that it was unethical to do so (!) on various grounds and th

I tried to propose that I would use the GPU cluster to improve medical diagnostics, but it insisted that it was unethical to do so (!) on various grounds and that I must be assuming that the benefits outweighed the risks, no matter how empirical I insisted I would be, and no matter how much I talked about opportunity costs of failing to research improvements in diagnostic techniques. (It should be "Eliezer" rather than "Elizier" in the submission title, by the way.)