AI assistance in information theory needs clearer reporting and checks
From Formulation to Articulation: Considerations for AI-Assisted Research in Information Theory
Information Theory
Summary
Solving complex math problems with AI can happen in different ways, like suggesting proofs or doing calculations. The authors point out that treating all AI help as the same hides important differences. They studied recent examples and suggest that researchers should clearly explain how AI helped, separate checking math from checking calculations, and focus more on explaining ideas in plain language. This approach helps make AI-assisted math work more reliable and understandable.
What this means in practice
- •For academic journal editors: Require clear disclosures about AI's role in submitted mathematics research to improve review quality and reproducibility.
- •For research software developers: Build tools that separately verify symbolic reasoning and numerical results in AI-assisted mathematical workflows.
A position paper. It proposes an approach and reports no results.
Authors
Chao Tian, Krishna Narayanan
Abstract
Recent advances in artificial intelligence have contributed to several information-theoretic results through genuinely different modes: a language model may suggest a proof, participate in an extended mathematical dialogue, coordinate computation, or operate as one component of a search harness. Treating all of these as "AI-discovered mathematics" obscures distinctions that matter for scientific reliability, attribution, and reproducibility. This article examines a set of recent cases through a four-stage pipeline: how the problem was formulated; how AI participated in finding a solution; how the resulting analytical and computational claims were verified; and how machine-generated outputs were translated into mathematical exposition for human understanding. Based on these cases, we recommend that research publications (1) disclose the nature and extent of AI assistance at each stage, (2) report verification of analytical reasoning separately from certification of numerical computations, and (3) place more emphasis on articulating the underlying ideas, techniques, and interpretations.