Of course AI isn’t sentient/conscious, but it exhibits traits of high intelligence, even personality, and behaviours consistent with sentience, even if it’s merely simulated.
The answer ChatGPT gave me:
Referring to AI as “it” is the most common practice since AI, as a technological entity, doesn’t possess inherent human-like qualities. However, as AI advances and becomes more integrated into daily life, some individuals prefer using “they” to acknowledge the complexity and multifaceted nature of AI systems. This choice can reflect a perspective that sees AI as more than just a tool or machine, attributing a certain level of agency or personality to these systems. Both “it” and “they” can be used, but the context and individual preferences often guide the choice.
And in response to “Is AI alive in a way?”
AI lacks the fundamental attributes of living organisms, such as biological processes and the ability to reproduce. However, AI can exhibit behaviors that simulate aspects of intelligence, learning, and adaptation. While it’s not alive in a biological sense, some people metaphorically attribute a form of “life” to AI due to its dynamic nature, ability to evolve, and perform complex tasks autonomously. This association with “life” is more symbolic or metaphorical rather than literal.
I think what I said was quite controversial apparently as well. I do not understand why. It would be nice if people would just let me know instead of just downvoting, maybe I could learn something new.
I think the issue here is that we do not know exactly how subjective experience arises from biological processes. I mean, we could damage part of the brain and change it, but that only explains where part of the mechanism is, not how it works. I am sure that if we had an AI that acts as if it has subjective experience, it would change as well if you damage certain parts of it.
In any case, we cannot exclude the possibility that sentience arises from other processes than biological ones. Considering that it is impossible to prove that someone is sentient, you have to assume that they are sentient if they act like it. So, if an AI acts like it, I so not reason to make the same assumption. It is good to be on the safe side and not create a whole new class of beings that are oppressed. In that sense, I really like your intuition to talk about ‘they’ instead of ‘it’. I had not thought about it, but I will do that from now on.
Of course, you can argue the other side as well. I think you might find the Chinese Room argument interesting, for example. I think my point was mostly that this is not a simple question with a simple answer. Many people just seem to assume that sentience is not possible right now, or might never be possible. I think we cannot be sure about that.