The recently proposed O(n log n) integer multiplication algorithm [pdf]

The discussion revolves around a recently claimed o(n log n) integer multiplication algorithm and reactions to its correctness, readability, and implications. Several commenters (jdb1729 and platypii) report a Lean formalisation of the result and ongoing improvements to exponents, and jdb1729 also points out the new result does not immediately improve polynomial multiplication over fixed finite fields. Some participants (flebron) stressed the technical distinction between o(n log n) and O(n log n). Reactions to the papers’ exposition split: yaoke259 and ajkjk praised the prose and candor, while Kapura and kenferry said the “maths dump” is hard to parse; kenferry and others found that the LLM Fable could read and explain the papers far faster than humans, and DannyBee proposed a concrete combinatorial argument clarifying one of the technical claims.

Opinion divides sharply over reliance on AI and practical impact. A number of commenters (kenferry, krackers, jrflo) accept LLMs as useful tools to accelerate understanding and validation, whereas others (Kapura, flockonus, khazhoux, throwaway63467) warned that opaque AI endorsements could produce false confidence or hallucinations. Several people (Eji1700, atzeus, autuni) emphasized this is mainly a theoretical advance unlikely to change everyday CPU integer multiply implementations immediately. Community efforts to build trust via formal proofs in Lean and crowdsourced formalisation (marwahaha, jdb1729) were presented as constructive paths toward verification.