I think there are two issues bundled into your reply, and they're best addressed separately. The small issue is whether people are responsible for what they pu

I think there are two issues bundled into your reply, and they're best addressed separately. The small issue is whether people are responsible for what they publish under their own name. Seems like a straightforward "yes", and whatever helper tools they use are irrelevant. The much bigger issue is why scientific publishing's standard for a review is only "plausible and human-like", allowing people to to submit a LLM generated summary of an abstract without fear of responsibility.