They're parodying those who claim that AGI will never exceed human intelligence.

IMO AGI will never exist period let alone exceed human intelligence. Sure we'll have increasingly powerful pattern recognition, but that's all it will ever be

And that’s fine, you might even be right. It’s not the opinion they are parodying, it’s the foolish arguments often used to “prove” it.

This was posted today, I assume there is some relation - https://news.ycombinator.com/item?id=29289431

Unfortunately, for the authors, this was not posted on April 1st.

Arguments that superintelligent machines are impossible.

Much funnier now that it's been explained to me. I certainly lack even human intelligence at some moments.

Indeed. I hope posting jokey papers on the Arxiv does not become a thing.

Seems as superficial and facile as saying machines will never wear hats.