A simplified proof of ETH is False [pdf]

The paper’s title confused one commenter, who read it as claiming to simplify a proof that the Exponential Time Hypothesis is false, rather than arguing against the hypothesis itself. Others questioned the work’s reliability because it appeared within hours of the original result and asked whether AI had produced it. The paper’s disclosure says the author used Claude and ChatGPT to simplify proofs, then manually expanded and explained them. One commenter worried that AI-generated mathematical work would be fed back into the models that produced it.

The debate also turned to the author’s credibility. One commenter accused him of presenting inflated credentials and misrepresenting a Princeton connection, citing his website and inconsistencies in linked papers. Others pushed back: one said he had been a Princeton lecturer or teaching assistant, while another pointed to papers and citations predating the current work. A commenter noted that a Princeton faculty search did not list him, while another argued that checking Google Scholar would have quickly shown a record of earlier cited work. Thus, commenters disagreed both about the author’s credentials and about whether AI assistance made the proof suspect.