I’ve been of the opinion that if we can’t tell the difference between an AI that mimics consciousness and an AI that is conscious, then we should treat them both as if they were. I don’t believe that we will create a conscious AI, (not that I’d be willing to die on this hill) but even if we create is something that mimics it, we should not deny it the rights of a conscious being.

I think it’s important that we start the conversation now, while AI is still largely unconvincing, well before we cross any threshold towards consciousness or convincing mimicry.