Originally Posted by Locobot

Nothing in your hypothetical "animal testing" example fits the criteria of malicious harm. I'm even fine with experiments with humans that result in grievous bodily harm and even death if the experiment is conducted ethically. Doctors who study medicines for terminal illnesses often knowingly give their patients a placebo instead of anything that might cure them. The difference is that those patients give their consent and know that the medicine they receive might be a sugar pill. If their consent is a result of coersion then the experiment is unethical and therefore unvalid.

Of course ethical standards are a highly charged political subject, but they must be in place in order to gather scientific data. Some ethical standards come from our laws, but the vast majority comes from the scientific community itself. Scientists decide the validity of others' scientific work, this applies to all methodology including ethics.