5 ms·

This is silly. The whole history of humankind is about reducing suffering, of ourselves and those around us. That's all we do. Also, suffering is a bug (or a f

by Geee 3y ago

This is silly. The whole history of humankind is about reducing suffering, of ourselves and those around us. That's all we do. Also, suffering is a bug (or a feature) of biological organisms. Obviously machines shouldn't be engineered to be capable of suffering.

ly3xqhl8g9 3y ago
The conundrum appears when the machine needs to suffer in order to become an agent taking decisions in our world. For example, the machine should know not to cause a fire in my house while doing the cleaning, it should be "afraid" of fire. My point is, if synthetic agents do indeed need to suffer to be agents, we will accept it just as easily we accept now that our cars take on rust. And of course, we will minimize the agent's suffering accordingly: no need for a robot folding clothes to experience existential dread as if they read Albert Camus, just be "afraid" of tousled clothes.
- Geee 3y ago
  We should be able to send robots into a fire, without a second thought. The robot should not feel any kind of suffering from the fire. Whoever owns the robot should decide how the robot should act. It may calculate the expected profit / cost of the action for the owner, but the owner should make the final decision. Programming it to avoid setting things on fire is just programming; there's no need for emotions or other complex biological rewards functions which are hard to replicate in computers. Engineering an agentic AI or a robot, i.e. a machine with independent will, is not a good idea at all, and shouldn't be pursued.
  - ly3xqhl8g9 3y ago
    We of course don't know how to engineer agency, but if we want black box top-down persuasion, how we train a dog without knowing how each neuron in the dog's brain works, some independent will may be required: you don't want to tell the robot how to fight each fire specifically (or train it on petabytes of fires just to have it stop in front of a shop called STOP [1]), you want to prove some general guidelines, and let it handle the actual cases itself. [1] Don't name your Store "Stop", https://www.reddit.com/r/SelfDrivingCars/comments/k56za3/dont_name_your_store_stop/ https://www.reddit.com/r/SelfDrivingCars/comments/k56za3/don...

by Geee 3y ago

ly3xqhl8g9 3y ago

The conundrum appears when the machine needs to suffer in order to become an agent taking decisions in our world. For example, the machine should know not to cause a fire in my house while doing the cleaning, it should be "afraid" of fire. My point is, if synthetic agents do indeed need to suffer to be agents, we will accept it just as easily we accept now that our cars take on rust. And of course, we will minimize the agent's suffering accordingly: no need for a robot folding clothes to experience existential dread as if they read Albert Camus, just be "afraid" of tousled clothes.

Geee 3y ago
We should be able to send robots into a fire, without a second thought. The robot should not feel any kind of suffering from the fire. Whoever owns the robot should decide how the robot should act. It may calculate the expected profit / cost of the action for the owner, but the owner should make the final decision. Programming it to avoid setting things on fire is just programming; there's no need for emotions or other complex biological rewards functions which are hard to replicate in computers. Engineering an agentic AI or a robot, i.e. a machine with independent will, is not a good idea at all, and shouldn't be pursued.
- ly3xqhl8g9 3y ago
  We of course don't know how to engineer agency, but if we want black box top-down persuasion, how we train a dog without knowing how each neuron in the dog's brain works, some independent will may be required: you don't want to tell the robot how to fight each fire specifically (or train it on petabytes of fires just to have it stop in front of a shop called STOP [1]), you want to prove some general guidelines, and let it handle the actual cases itself. [1] Don't name your Store "Stop", https://www.reddit.com/r/SelfDrivingCars/comments/k56za3/dont_name_your_store_stop/ https://www.reddit.com/r/SelfDrivingCars/comments/k56za3/don...