This kind of thing, much more than LLMs, makes me worry about AGI takeoff.
My meek opinion is this is obvious. Human-level intelligence requires at most 20 watts and substrate no more complicated than can be constructed from simple organic molecules in a dirty environment.
What is possible with 20 kilowatts and wafer fabricators?
Anyway, OP is about lossy compression. I can't fully follow it but they talk about techniques for mitigating loss later in the paper.