这种治理会形成选择性的数学本体。Lean 对数论与代数几何较为适配，但对图论与范畴论则不那么顺手。Dick 与其他人指出，实作选择能微妙地重导数学优先序：符号与系统限制可能将研究者推向更易形式化的概念，而非纯粹以问题直觉推进。于是，注意力可能从数学问题本身转向形式系统的行为，进而改变「自然」数学理解。
文中将这一现象称为对 rigor 的「过度校正」。Akshay Venkatesh 将现代的形式优雅与早期证明对照，称早期论证虽较粗粝，却更容易被人类直觉把握；他认同证明是核心，但质疑证明是否应成为数学的定义性特征。Asok 所言有研究者主张唯有将论文 Lean-formalize 才是前进道路，但另一种主张是退一步、少写一点，惟既有奖励机制使其难以实现。文章最后指出，数学具有历史性的自我修正力；本文于 2026 年 3 月 25 日发表，并在 3 月 27 日更正将 Maurice Mashaal 厘清为记者而非数学家。
The article argues that while Lean has accelerated the digitization of mathematical proof, it also imports governance problems. In 1994 the QED Manifesto collapsed into disputes over language, definitions, and frameworks, and similar disagreements resurfaced historically. Lean now relies on a centralized community process: users decide which definitions enter the library and how they are encoded, much like Wikipedia. Emily Riehl describes this as broadly democratic, but still warns that a single canonical choice must eventually be made even when multiple mathematically valid options exist.
That governance creates a selective ontology. Lean aligns well with areas such as number theory and algebraic geometry, but is less suited to graph theory and category theory. Dick and others argue that implementation choices can subtly redirect mathematical priorities: notation and system constraints may nudge researchers toward concepts that formalize cleanly, away from problem-intuitive development. He observes that this can shift attention from the mathematical domain itself toward the behavior of the formal system, changing what counts as natural mathematical understanding.
The piece frames this as “overcorrection” in rigor. Akshay Venkatesh contrasts modern formal elegance with earlier proof styles, calling older arguments gritty yet cognitively graspable. He accepts proof as central yet questions whether proof should remain math’s defining feature. Asok reportedly says some communities now treat Lean-formalization as the only path, while alternative voices urge stepping back, although publishing incentives discourage that. The article ends by stressing math’s historical self-correction, noting a March 25, 2026 publication and a March 27 correction that reclassified Maurice Mashaal as a journalist, not a mathematician.