Again, my concern is about the legal implications of declaring sentience when there is no way to verify it, but there is every guarantee of it being misused and abused by the super privileged to strip mine the planet and the society. In the future when the AI is assumed to have achieved sentience, will it be able to demand that it should not be used as a tool to exploit the rest of the humanity? If so, go ahead and give it rights. But somehow, that's not how I expect it to end. Late stage capitalism has the dubious record of perverting every technology it can get its hands on.

Even now, AI is hardly the panacea that is saving humanity from an impending crisis. Far from it, it's currently the vessel for copyright-washing (and gpl-washing) creative work, wrecking the job market based on the rhetoric of abilities that are well beyond the reality, the machine to do things that we were already doing (like web searches) at 10x the energy costs, and run massive data centers that produce noises well above the hazard levels and CO2 emissions equivalent to that of small nations.

I understand that AI technology is capable of novel applications and in fields that need imminent attention (like climate research). But that's not what drives it today - it's profit seeking bordering on insanity. They're getting people addicted to uses that are hardly novel and comes at a steep energy cost. Why else are these companies so worried about AI rights when they have scant regards for human rights? I'm not concerned about what AI can do. I'm concerned about what AI will do.