[colmvp]

Your statement made me realize I should empathize with the machines in the Matrix. After all, a human life is objectively worth less to machines than to humans. A machine can last for centuries and can have more utility on a daily basis than a human.

[stametseater]

Different things have different values, which you seem to have realized here: "a human life is objectively worth less to machines" "To machines" is the important part. But unless you're a machine, it would be strange for you to have machine values, so I don't understand why you say that you should empathize with the machines. Unless... are you an LLM?

Anyway, different sorts of things having different values is why so many people are concerned about powerful non-human entities with inhuman motivations (AIs generally, and corporations particularly.) It's also the reason people are afraid of hungry bears (the bear cares more about itself than you.)

[optymizer]

I wasn't arguing utility. I'm arguing there there's some value we place on human life that is above a dog's life. It doesn't depend on how much utility one derives from humans vs dogs.

The value judgement could simply come from the fact that we need to preserve our own species, and if we need to sacrifice some small percentage of dogs to do so, than this is perfectly reasonable. We can still exhibit empathy by accepting that this situation is not ideal and striving to reduce that percentage down to 0, maybe by inventing substitutes.

As for the machines example, I would expect the machines to place higher value on their own life than on human life, precisely because they would be objectively superior, which is why I would not let machines exist without any restrictions, even if the machines would be a more useful dominant species.

[nottathrowaway3]

Utility (= (pleasure minus pain) x time) is the essential good when we're talking about animal and human welfare. Good actions increase expected utility; bad actions decrease expected utility.

Machines experience no pleasure or pain and thus there is zero utility gained/lost by destroying them. Dogs and humans do; it's bad to kill or torture them.

Throwing a dog in front of a train (or killing/torturing it for research) to save a child depends on how likely it is that it will work, and how many QALYs the dog and child have left.